Skip to main content

Full text of "Journal of the Society of Motion Picture Engineers"

See other formats


From  the  collection  of  the 


Prefinger 

i     a 

Uibrary 


San  Francisco,  California 
2007 


JOURNAL 

OF  THE  SOCIETY  OF 

MOTION  PICTURE  ENGINEERS 

Volume  XXXII  January,  1939 

CONTENTS 

Page 
Undersea  Cinematography E.  R.  F.  JOHNSON      3 

The  Road  Ahead  for  Television I.  J.  KAAR     18 

Report  of  the  Studio  Lighting  Committee 44 

Photographic  Effects  in  the  Feature  Production  "Topper" 

R.  SEA  WRIGHT  AND  W.  V.  DRAPER     60 

Latent  Image  Theory  and  Its  Experimental  Application  to 
Motion  Picture  Sound-Film  Emulsion W.  J.  ALBERSHEIM     73 

The  Evaluation  of  Motion  Picture  Films  by  Semimicro  Testing 

J.  E.  GIBSON  AND  C.  G.  WEBER  105 

Current  Motion  Picture  Literature 110 

Spring,  1939,  Convention '. 113 

Society  Announcements 117 


JOURNAL 

OF  THE  SOCIETY  OF 

MOTION  PICTURE  ENGINEERS 


SYLVAN  HARRIS,  EDITOR 
Board  of  Editors 

J.  I.  CRABTREE,  Chairman 

A.  N.  GOLDSMITH  L.  A.  JONES  H.  G.  KNOX 

A.  C.  HARDY  E.  W.  KELLOGG  G.  E.  MATTHEWS 


Subscription  to  non-members,  $8.00  per  annum;  to  members,  $5.00  per  annum, 
included  in  their  annual  membership  dues;  single  copies,  $1.00.  A  discount 
on  subscription  or  single  copies  of  15  per  cent  is  allowed  to  accredited  agencies. 
Order  from  the  Society  of  Motion  Picture  Engineers,  Inc.,  20th  and  Northampton 
Sts.,  Easton,  Pa.,  or  Hotel  Pennsylvania,  New  York,  N.  Y. 
Published  monthly  at  Easton,  Pa.,  by  the  Society  of  Motion  Picture  Engineers. 

Publication  Office,  20th  &  Northampton  Sts.,  Easton,  Pa. 
General  and  Editorial  Office,  Hotel  Pennsylvania,  New  York,  N.  Y. 

West-Coast  Office,  Suite  226,  Equitable  Bldg.,  Hollywood,  Calif. 
Entered  as  second  class  matter  January  15,  1930,  at  the  Post  Office  at  Easton, 
Pa.,  under  the  Act  of  March  3,  1879.     Copyrighted,  1939,  by  the  Society  of 
Motion  Picture  Engineers,  Inc. 

Papers  appearing  in  this  Journal  may  be  reprinted,  abstracted,  or  abridged 
provided  credit  is  given  to  the  Journal  of  the  Society  of  Motion  Picture  Engineers 
and  to  the  author,  or  authors,  of  the  papers  in  question.  Exact  reference  as  to 
the  volume,  number,  and  page  of  the  Journal  must  be  given.  The  Society  is  not 
responsible  for  statements  made  by  authors. 


OFFICERS  OF  THE  SOCIETY 

**  President:     E.  A.  WILLIFORD,  30  East  42nd  St.,  New  York,  N.  Y. 
**  Past-President:     S.  K.  WOLF,  RKO  Building,  New  York,  N.  Y. 
**  Executive  Vice-President:     N.  Levinson,  Burbank,  Calif. 

*  Engineering  Vice-President:     L.  A.  JONES,  Kodak  Park,  Rochester,  N.  Y. 
**  Editorial  Vice-President:    J.  I.  CRABTREE,  Kodak  Park,  Rochester,  N.  Y. 

*  Financial  Vice-President:    A.  S.  DICKINSON,  28  W.  44th  St.,  New  York,  N.  Y. 
**  Convention  Vice-President:    W.  C.  Kunzmann,  Box  6087,  Cleveland,  Ohio. 

*  Secretary:    J.  FRANK,  JR.,  90  Gold  St.,  New  York,  N.  Y. 

*  Treasurer:    L.  W.  DAVEE,  153  Westervelt  Ave.,  Tenafly,  N.  Y. 

GOVERNORS 

**  M.  C.  BATSEL,  Front  and  Market  Sts.,  Camden,  N.  J. 

*  R.  E.  FARNHAM,  Nela  Park,  Cleveland,  Ohio. 

*  H.  GRIFFIN,  90  Gold  St.,  New  York,  N.  Y. 

*  L.  L.  RYDER,  5451  Marathon  St.,  Hollywood,  Calif. 

*  A.  C.  HARDY,  Massachusetts  Institute  of  Technology,  Cambridge,  Mass. 

*  S.  A.  LUKES,  6427  Sheridan  Rd.,  Chicago,  111. 

**  H.  G.  TASKER,  14065  Valley  Vista  Blvd.,  Van  Nuys,  Calif. 


*  Term  expires  December  31,  1939. 
**  Term  expires  December  31,  1940. 


UNDERSEA  CINEMATOGRAPHY 
E.  R.  F.  JOHNSON** 


Summary. — The  dates  of  the  first  recorded  use  of  underwater  photography  and 
the  tendencies  toward  its  increasing  use  by  producers  are  noted,  and  the  author's 
early  experiences  in  this  field  are  described.  For  work  in  natural  settings  the  most 
useful  equipment  consists  of  submergible  cameras  placed  on  the  bottom  and  operated 
by  divers.  The  problems  of  and  equipment  for  such  work  are  dealt  with  and  it  is 
pointed  out  that  studio  tank  work  shares  most  of  these  problems. 

The  optical  properties  of  water  are  described.  Since  water  is  less  transparent 
than  air,  photography  by  natural  light  is  limited  to  shallow  depths  and  more  power 
is  required  for  artificial  illumination  under  water.  Since  colors  are  not  absorbed 
equally,  accurate  monochrome  rendering  and  photography  in  natural  color  are  com- 
plicated. Haze  limits  the  distance  at  which  pictures  can  be  taken  under  water,  but 
is  largely  confined  to  a  part  of  the  spectrum  and  can  be  partially  eliminated  by  the 
use  of  color  filters.  It  is  plane  polarized  and  can,  therefore,  also  be  suppressed  by 
the  use  of  polarizing  plates.  The  advantages  of  this  method  are  briefly  stated — it 
does  not  distort  monochrome  rendering  and  can  be  used  in  natural  color  photography. 

The  ideal  attributes  of  equipment  for  use  in  underwater  cinematography  are  outlined 
and  available  equipment  is  briefly  described. 

HISTORY  AND  CURRENT  STATUS 

The  first  recorded  attempt  to  take  photographs  under  water  that 
has  come  to  our  attention  was  by  Boutan  in  1893  and  we  understand 
that  he  succeeded  in  securing  a  few  fairly  successful  still  pictures. 

The  possibilities  of  underwater  motion  pictures  seem  to  have  in- 
trigued the  fancy  of  commercial  producers  almost  as  early,  if  indeed 
not  earlier,  than  it  did  the  scientists  and  educators.  Williamson 
produced  the  underwater  picture,  Twenty  Thousand  Leagues  under  the 
Sea,  in  1915-16,  and  at  about  the  same  time  scientists,  among  whom 
were  Bartsch,  Beebe,  and  Minor,  started  using  water-tight  motion 
picture  camera  housings  in  conjunction  with  the  suitless  type  of 
diving  helmet,  in  order  to  take  motion  pictures  with  which  to  illustrate 
their  lectures  upon  underwater  life. 

*  Presented  at  the  1938  Fall  Meeting  at  Detroit,  Mich. ;   received  October  3, 
1938. 

**  Mechanical  Improvements  Corporation,  Moorestown,  N.  J. 

3 


4  E.  R.  F.  JOHNSON  [j.  s.  M.  P.  E. 

Recently  the  tendency  of  producers  to  show  what  happens  under 
water  as  a  part  of  their  stories  has  grown  vastly,  to  say  nothing  of 
pictures  having  the  principal  parts  of  their  plots  based  on  action 
allegedly  taking  place  there.  Indeed  pictures  of  champion  divers  or 
swimmers  are  no  longer  considered  complete  without  a  view  of  their 
graceful  evolutions  after  penetrating  the  surface.  Some  real  ocean 
water  scenes  were  used  in  the  excellent  story  Submarine  DI  and 
underwater  scenes  can  be  used  to  add  to  the  romantic  touch  of  a 
picture,  as  was  done  in  Jungle  Love.  The  work  of  naval  and  com- 
mercial divers  has  as  yet  hardly  been  touched  upon  and  their  heroic 
exploits  offer  material  for  a  host  of  future  thrillers.  People  in  ever- 
increasing  numbers  are  becoming  cognizant  of  the  real  underwater 
conditions.  For  instance,  Miami  University  has  a  class  in  marine 
zoology  where  the  students,  using  diving  helmets,  go  below  the  sur- 
face. At  the  Marine  Studios  in  Florida  and  at  the  Bermuda  Aquarium 
tourists  put  on  diving  helmets,  or  observe  the  underwater  world 
through  ports.  In  France,  Paul  Painleve's  underwater  club  is  edu- 
cating another  section  of  the  public.  This  increasing  familiarity 
is  making  audiences  more  critical,  and  the  technic  and  equipment  for 
actual  underwater  photography  as  contrasted  to  shots  through  glass 
port-holes  in  tanks  are  of  both  present  and  growing  importance  to 
entertainment  pictures,  as  well  as  to  the  scientist  and  educator. 

PHOTOGRAPHIC  CONDITIONS  UNDER  WATER 

The  author's  vigorous  attack  on  the  problems  involved  in  under- 
water photography  was  brought  about  by  a  stinging  defeat  in  1928. 
We  read  one  of  Beebe's  glorious  descriptions  of  the  beauties  of  under- 
sea gardens  and  the  complete  ease  with  which  they  could  be  visited 
and  photographed.  An  Eyemo  camera  was  enclosed  in  a  simple  case 
with  a  box  of  calcium  chloride  to  keep  the  condensation  off  the  lens 
and  window.  The  result  of  several  weeks'  work  was  very  mediocre, 
however,  for  we  found  out  that  if  one  can  see  forty  feet  that  does  not 
mean  that  he  can  take  good  pictures  at  more  than  ten,  or  always 
even  up  to  ten.  We  found  that  natural  light  is  strong  enough  for 
photography  under  water  only  between  10 : 30  A.M.  and  3 : 30  P.M.  in 
the  summer  and  even  less  in  winter ;  that  the  pellucid  tropic  seas  are 
more  often  than  not  full  of  white  sand,  green  algae,  gray-green  marl, 
or  other  detritus;  and  that  if  the  undersea  photographer  hoped  to 
get  anything  in  a  natural  set  he  had  to  be  lightning  fast  to  grasp  his 
opportunity. 


Jan.,  1939]  UNDERSEA  CINEMATOGRAPHY  5 

Diving  bells  and  baby  submarines  were  considered,  but  after 
talking  with  a  few  persons  who  had  had  experience  with  such  equip- 
ment, we  saw  that  they  could  not  be  transported  and  put  into  position 
with  sufficient  ease  and  speed  to  suit  the  underwater  photographer, 
and  they  are  the  plaything  of  every  wave  or  squall  of  wind.  We  at- 
tached cameras  to  water  telescopes;  we  shot  them  through  glass- 
bottomed  buckets;  we  submerged  them  and  sighted  through  peri- 
scopes. But  pictures  from  an  unsteady  base  are  unattractive  and 
tend  to  make  the  audience  seasick.  It  is  our  opinion,  therefore,  that 
underwater  pictures  can  best  be  made  from  the  bottom  and  not  the 
surface ;  also,  that  a  compact  underwater  camera  operated  by  a  diver 
should  be  used  for  all  picture  work  in  the  open  sea,  lakes,  and  rivers. 
The  same  applies  to  a  considerable  degree  also  in  swimming  pools  and 
tanks,  for  the  cameraman  shooting  through  a  port-hole  is  greatly 
hampered  in  following  a  moving  object,  and  most  underwater  sub- 
jects depend  on  action  rather  than  expression  to  tell  their  story. 

The  physical  qualities  of  water  are  responsible  for  many  of  the 
difficulties  encountered  in  underwater  photography.  The  purest  of 
water  is  far  less  transparent  than  air;  and,  because  it  is  an  excellent 
solvent,  it  is  rarely  pure  in  nature;  and  dissolved  matter  profoundly 
affects  the  optical  properties.  Then,  too,  water,  having  greater  den- 
sity and  viscosity  than  air,  supports  a  much  greater  proportion  of 
suspended  matter  both  organic  and  inorganic,  and  this  has  an  even 
greater  effect  on  its  optical  properties.  The  quantity  and  kind  of 
dissolved  matter  are  relatively  constant  for  any  location  and,  at 
least  in  sea  water,  are  nearly  the  same  for  almost  all  areas  where 
underwater  pictures  can  be  taken.  Suspended  matter,  on  the  other 
hand,  is  highly  variable  both  for  different  locations  and  at  the  same 
location  with  varying  season  and  weather.  It  is  the  quantity  of 
suspended  organisms  and  particles  that  finally  determine  whether 
or  not  satisfactory  pictures  can  be  taken  at  a  particular  place  or  time. 

OPTICS  OF  UNDERWATER  PHOTOGRAPHY 

In  undersea  photography  it  is  general  practice  to  use  ordinary  cine 
lenses  computed  for  use  in  air — protected  by  a  plane  window.  This 
introduces  a  water-air  boundary  which  affects  the  focus  and  correc- 
tions of  the  lens.  Objects  under  water  appear  nearer  and  larger  both 
to  the  eye  and  to  a  camera.  We  have  computed  the  effect  upon  focus 
and  it  turns  out  that  the  ratio  of  the  air  focus  to  that  under  water  is 
equal  to  the  index  of  refraction  of  air  with  respect  to  water.  The 


6 


E.  R.  F.  JOHNSON 


[J.  S.  M.  P.  E 


index  varies  with  the  salinity  and  temperature  of  the  water  but  the 
value  0.750  may  be  used  for  all  conditions  with  negligible  error.  It 
follows  that  to  focus  on  an  object  at  any  distance  under  water  the 
same  lens  extension  is  required  as  for  an  object  at  three-quarters  of 
that  distance  in  air. 

The  presence  of  the  water-air  boundary  in  front  of  the  lens  also 
introduces  both  spherical  and  chromatic  aberration.  Fortunately, 
if  the  plane  of  the  window  is  perpendicular  to  the  axis  of  the  lens, 
they  are  both  too  small  to  require  correction.  In  tank  work  any 
attempt  to  position  a  camera  other  than  normal  to  the  plane  of  the 
window  will  result  in  objectionable  aberration. 


5000  6000 

VIOLET  Bi_ue  GREEN         YELLOW       OKAVKE 

FIG.  1.     Spectral  quality  of  mean  noon  sunlight  at  the 
surface,  and  through  25  and  50  feet  of  water. 

When  photographing  by  sunlight,  the  first  factor  to  be  considered 
is  the  overall  reduction  in  intensity  of  light.  This  varies  greatly  with 
conditions,  but  it  is  our  experience  that  under  average  conditions 
twenty-five  to  thirty-five  feet  is  the  limiting  depth  using  rapid  lenses 
and  Eastman  Super  X  or  Agfa  Supreme  film  and  a  filter  with  a  factor 
of  from  two  to  four.  The  newer  ultra-rapid  films  extend  this  limit 
somewhat.  Fortunately  the  largest  percentage  of  interesting  marine 
life  and  human  activity  is  to  be  found  within  this  range.  At  greater 
depths  photographic  subjects  become  more  scarce  and  difficulties  are 
materially  increased. 

A  further  complication  is  added  by  the  fact  that  water  does  not 
absorb  different  colors  equally.  Fig.  1  shows  the  approximate  inten- 
sity and  spectral  quality  of  sunlight  at  the  surface  and  at  several 
depths  in  water.  The  curves  were  computed  from  the  laboratory 


Jan.,  1939] 


UNDERSEA  CINEMATOGRAPHY 


measurements  of  the  transparency  of  sea  water  by  E.  O.  Hulbert1  and 
approximate  the  conditions  found  in  practice.  It  will  be  seen  that 
sea  water  is  most  transparent  in  the  blue-green  region  between  4400 
and  5400  A  and  that  the  red  is  quickly  absorbed.  This  filtering  ac- 
tion of  water  makes  difficult  a  true  monochrome  rendering  of  subjects, 
and  has  an  even  greater  effect  on  photography  in  natural  colors.  In 
color  photography  compensating  filters  can  be  used  to  correct  for 
the  spectral  quality  of  the  light  at  any  given  depth.  It  is  to  be  noted, 
however,  that  theoretically  a  different  filter  would  be  required  for 
every  depth.  Moreover,  the  same  would  be  true  for  different  dis- 
tances from  the  camera  to  the  object.  Thus,  if  an  object  at  six  feet 


ZS 


4000 


S5oo 


4500  5000 

WAVE      LENGTH      A 

FIG.  2.     Spectral  distribution  of  water  haze  in  sea  water; 
transmission  curve  of  Wratten  Aero  No.  2  filter. 


from  the  camera  and  six  feet  deep  is  being  photographed,  a  compen- 
sating filter  correct  for  twelve  feet  of  water  would  be  required.  How- 
ever, even  though  filters  are  used  objects  closer  than  six  feet  would 
tend  to  be  too  red  and  at  greater  than  six  feet  would  be  progressively 
greener — the  background  fading  out  in  a  uniform  blue-green.  This, 
in  fact,  is  a  real  effect.  Objects  at  a  distance  do  not  appear  to  be  the 
same  color  to  a  diver  as  when  they  are  close  by.  A  diver's  vision 
fades  out  in  a  misty  blue-green  haze.  Color-film,  however,  accen- 
tuates this  effect,  making  the  background  an  unnaturally  intense 
blue-green.  We  have  taken  both  Kodachrome  and  Duf  ay  color  stills 
and  Dufaycolor  motion  pictures,  most  of  which  exhibit  this  effect. 
By  proper  limitation  of  depth  and  distance,  however,  beautiful 
results  can  be  obtained. 


8  E.  R.  F.  JOHNSON  [J.  S.  M.  P.  E. 

The  greatest  bete  noire  of  the  underwater  photographer  is  water 
haze  or  "nuisance  light."  It  is  strictly  analogous  to  aerial  haze  but 
being  much  more  intense  its  effect  shows  up  in  the  picture  of  an  ob- 
ject only  a  few  feet  away  rather  than  a  matter  of  miles.  This  haze 
originates  in  the  scattering  of  light  by  the  water  and  by  dissolved  and 
suspended  matter  between  the  camera  and  the  object.  Its  effect  is  to 
cause  a  uniform  exposure  over  the  whole  picture,  which  tends  to  mask 
detail  and  contrast;  as  the  distance  becomes  greater  the  haze  becomes 
brighter,  compared  to  the  brightness  of  the  object,  finally  masking  it 
completely. 

It  was  felt  that  water  haze,  like  aerial  haze,  should  consist  prin- 
cipally of  light  in  a  limited  spectral  region  and  that  a  color  filter 
would  eliminate  much  of  it.  With  this  in  mind  we  conducted  a 
series  of  experiments  with  an  underwater  spectrograph.  Fig.  2 
gives  the  relative  spectral  distribution  of  the  haze  light  in  the  sea 
water  off  the  Florida  Keys.2  Tank  tests  with  distilled  water  gave 
an  almost  identical  curve.  Camera  tests  showed  a  great  improve- 
ment when  a  Wratten  Aero  No.  2  filter  was  used.  The  transmission 
curve  of  this  filter  is  also  given.  Fig.  3  shows  the  improvement 
obtained  by  the  use  of  filters :  (a)  is  a  scene  at  a  distance  of  six  feet 
with  no  filter;  (b)  a  similar  scene  using  an  Aero  No.  2;  and  (c)  with 
a  Wratten  No.  25  filter,  which  transmits  only  the  red  rays  of  wave- 
lengths greater  than  6000  A.  This  last  picture  shows  only  very 
slight  improvement  in  detail  over  the  one  taken  with  the  Aero  No.  2, 
and  this  improvement  is  more  than  offset  by  the  unnatural  appear- 
ance of  the  subjects  and  by  the  extreme  exposure  increase  required. 
The  chief  objection  to  the  use  of  color  filters  to  eliminate  haze  is  the 
fact  that  water  is  most  transparent  to  the  blue-green  region  of 
the  spectrum ;  but  this  is  also  the  region  of  maximum  intensity  of  the 
haze  light,  so  in  eliminating  it  the  most  efficient  photographic  light 
is  also  lost.  Fortunately  there  is  another  means  of  cutting  out  this 
troublesome  haze. 

So  far  as  haze  light  is  concerned  studio  tank  work  offers  the  same 
problems  as  natural  settings — as  we  believe  at  least  one  producer  has 
found,  to  his  sorrow,  after  putting  several  thousand  gallons  of  expen- 
sive distilled  water  into  a  nicely  scrubbed  tank  and  then  failing  to  get 
the  clear  crisp  pictures  he  wanted  so  badly. 

The  fact  that  this  nuisance  light  is  present  even  in  distilled  water 
that  has  stood  long  enough  to  be  free  of  air  bubbles  indicates  that 
the  origin  of  much  of  it  must  be  molecular  scattering  by  the  water  it- 


Jan.,  1939] 


UNDERSEA  CINEMATOGRAPHY 


FIG.  3.  Use  of  color  filters  to  reduce  water 
haze;  (a)  no  filter.  (6)  Wratten  Aero  No.  2 
.filter,  (e)  Wratten  No.  25  filter. 


10  E.  R.  F.  JOHNSON  [j.  s.  M.  P.  E. 

self.  Therefore,  according  to  the  Raman-Einstein-Smolchowski 
theory  it  should  be  almost  completely  plane  polarized.3  Our  dis- 
covery of  this  fact  led  to  our  use  of  polarizing  screens.  By  the  use  of 
these  screens  it  is  possible  to  eliminate  a  greater  part  of  the  "nuisance 
light"  than  by  any  other  means.  Their  use  requires  an  exposure  in- 
crease of  from  two  to  four  times.  Unfortunately,  the  haze  light  is  not 
completely  polarized,  so,  while  the  distance  at  which  satisfactory 
pictures  can  be  obtained  is  extended,  there  is  still  a  very  definite  limit. 

Probably  the  most  advantageous  feature  of  this  method  of  elimi- 
nating haze  light  is  that  polarizing  screens  are  almost  perfectly  spec- 
trally neutral.  They  do  not  distort  the  monochrome  rendering  nor 
do  they  eliminate  the  most  useful  portion  of  the  spectrum  as  does  a 
yellow  or  red  filter.  This  spectral  neutrality  further  makes  possible 
haze  elimination  when  using  color-film.  However,  at  present  the 
speed  of  color-film  does  not  permit  the  use  of  both  a  compensating 
filter  and  a  polarizing  plate  under  normal  conditions.  If  the  speed 
of  these  materials  can  be  doubled  a  great  improvement  in  the  quality 
of  undersea  color  pictures  is  anticipated  by  the  use  of  polarizing 
screens. 

Light-rays  passing  into  water  through  the  surface  are  bent  until 
they  travel  nearly  straight  down,  so  by  natural  illumination  subjects 
under  water  are  inclined  to  be  overly  contrasty  with  highlights  on  top 
and  densely  shadowed  undersides.  We  tried  relieving  this  situation 
with  reflector  boards  but  found  that  boards  large  enough  to  help  at 
all  had  so  much  water  resistance  that  even  in  slack  water  they  were 
difficult  to  handle  and  with  any  current  it  became  impracticable 
either  to  set  them  or  keep  them  in  position.  Shadows  can  be  re- 
lieved to  a  certain  extent  by  the  use  of  artificial  lights.  Reflectors 
must  be  small  and  highly  efficient  or  they  become  unmanageable  in 
any  current.  In  using  lights,  it  is  necessary  to  exercise  extreme  care 
in  placement,  otherwise  haze  light  makes  the  lamp  beam  visible. 
Since  most  of  the  energy  from  incandescent  lights  is  in  the  red  end  of 
the  spectrum,  in  which  region  the  water  has  its  strongest  absorption, 
the  power  requirements  are  much  greater  than  for  equivalent  illumina- 
tion at  the  surface. 

EQUIPMENT 

Having  outlined  the  technical  and  practical  difficulties  confronting 
the  underwater  cinematographer  we  shall  now  briefly  describe  the 
ideal  attributes  of  apparatus  to  meet  these  difficulties  and  the  prac- 
tical equipment  developed  by/our  company,  for  this  specialized  field. 


Jan.,  1939]  UNDERSEA  CINEMATOGRAPHY  11 

Motion  picture  producers  early  learned  that  dependence  upon  na- 
tural light  could  be  extremely  expensive,  sometimes  tying  up  the 
whole  staff  of  actors  and  technicians  for  days  waiting  for  favorable 
conditions.  In  underwater  photography  the  clarity  of  the  water 
and  the  size  of  the  waves  are  also  factors  that  can  vary  independently 
of  sunlight,  thus  further  limiting  the  useful  part  of  time  on  location. 
Moreover,  even  under  the  best  of  conditions,  the  working  day  under 
sea  is  shorter  than  at  the  surface;  first,  because  the  sun  reaches  full 
brightness  later  and  fades  earlier;  and,  second,  long  exposure  to  water 
soon  tires  both  cameramen  and  actors.  These  factors  tend  to  run  up 
the  expense  of  underwater  footage  and  demand  a  high  standard  of  re- 
liability, convenience,  and  speed  of  operation  in  the  apparatus  used. 

Sending  equipment  to  the  surface  for  adjustments  of  stop  or  focus 
or  change  of  filter  is  wasteful  of  time.  It  is,  therefore,  the  first  re- 
quirement of  underwater  equipment  that  the  controls  for  all  adjust- 
ments be  quickly  and  conveniently  operable  under  sea  by  the  diver. 

The  fact  that  water  is  a  far  less  yielding  medium  than  air  dictates 
other  requirements.  A  camera  that  would  be  quite  stable  in  a 
twenty-mile  wind  might  easily  be  thrown  over  by  even  a  two-mile 
current.  When  working  at  small  depths  the  under-surface  surge 
from  waves  tends  to  sway  and  billow  both  diver  and  equipment. 
Very  often  the  diver  has  difficulty  in  standing  or  walking  even  with- 
out apparatus.  Therefore,  tripod  mount  and  spring  or  electric  motor 
drive  are  essential.  Apparatus  must  be  compact  to  keep  down  its 
water  resistance;  it  must  be  light  enough  for  ease  in  carrying,  yet 
heavy  enough  to  be  stable.  All  connections  must  be  rigid. 

When  under  water,  man  becomes  a  clumsy,  slow-moving  creature. 
It  is,  therefore,  important  that  any  couplings  that  must  be  made 
should  be  large  and  simple.  Calibration  markings  should  be  large 
and  distinct,  and  all  controls  and  calibrations  should  be  visible 
and  operable  from  one  position. 

Even  after  short  submergence  the  skin  of  a  diver's  hands  becomes 
softened  and  very  easily  cut  by  things  that  would  not  do  so  at  the  sur- 
face. For  this  reason  everything  must  be  smooth  with  no  sharp 
edges. 

The  construction  of  diving  helmets,  and  the  fact  that  the  camera- 
man has  difficulty  in  remaining  perfectly  still,  make  it  necessary  that 
view-finders  be  corrected  for  an  eye  position  well  back  of  the  port,  and, 
further,  the  reduced  illumination  makes  important  a  large  brilliant 
image. 


12 


E.  R.  F.  JOHNSON  [J.  S.  M.  p.  E. 


FIG.  4.     Professional  model  underwater  motion  picture   camera. 


FIG.  5.     Underwater  housing  for  Bell  &  Howell  Eyemo,  open  for  loading. 


Jan.,  1939]  UNDEkSEA  ClNEMAtOGkAPfiY  "13 

Direct  focusing  is  a  difficult  if  not  impossible  task  and  therefore 
accurate  focus  calibration  of  lenses  is  a  necessity. 

In  developing  apparatus  for  underwater  cinematography  we  have 
not  undertaken  the  design  of  new  camera  mechanisms  but  rather  have 
modified  for  underwater  use  the  excellent  existing  surface  equip- 
ment. The  materials  and  construction  of  all  apparatus  are  such  that 
no  condensation  occurs  on  windows  when  submerged,  thus  eliminating 
the  need  of  troublesome  and  time-consuming  chemical  driers.  In 
general,  it  may  be  said  that  any  surface  camera  can  be  encased  for 
such  use,  but  for  our  standard  models  we  have  selected  those  whose 
size  and  layout  make  them  most  adaptable  for  the  purpose. 


FIG.  6.     Professional  model  underwater  still  camera. 

The  most  complete  unit  is  the  professional  model  motion  picture 
camera,  Fig.  4,  which  was  designed  to  give  the  underwater  cinema- 
tographer  an  instrument  possessed  of  the  greatest  possible  flexibility 
and  convenience  of  operation. 

The  camera  mechanism  is  the  Akeley,  with  a  capacity  of  200  feet  of 
standard  35-mm.  motion  picture  film.  It  is  driven  by  an  electric 
motor  with  external  speed  control.  The  trigger  switch  is  in  a  sepa- 
rate water-tight  case  mounted  on  the  cable;  it  may  be  used  at  a  short 
distance  from  the  camera  or  mounted  on  the  guiding  handle  of  the 
tripod,  making  it  possible  to  guide  and  run  the  camera  with  the  left 
hand,  leaving  the  right  hand  free  to  adjust  focus  and  aperture.  A 


14  E.  R.  F.  JOHNSON  [j.  s.  M.  P.  E. 

Veeder  type  of  footage  counter  is  at  the  rear  of  the  case,  and  every  two 
feet  a  dim  light  flashes  at  one  side  of  the  view-finder,  permitting  the 
cameraman  to  keep  count  of  footage  without  looking  away  from  the 
scene.  Three  lenses,  wide-angle,  standard,  and  telephoto,  are 
mounted  in  the  instrument.  These  are  in  a  vertical  row  rather  than 
in  the  conventional  turret,  in  the  interest  of  simpler  lens  and  filter 
control.  Provision  is  made  for  two  color  niters,  which  may  be  thrown 
in  or  out  at  will,  and  a  polarizing  plate  which  may  be  racked  in  front 
of  the  lens  or  removed.  It  may  also  be  rotated  under  water  and  a 
sun's  position  indicator  on  the  rear  of  the  camera  indicates  the  proper 
rotation  when  using  natural  light. 

The  view-finder  gives  a  large  brilliant  image  and  incorporates  ad- 
justment for  correction  of  parallax.  Supplementary  lenses  are  used 

to  obtain  the  fields  of  the  dif- 
ferent camera  lenses  without 
a  reduction  in  the  image  size. 

In  air  the  camera  weighs 
approximately  seventy-five 
pounds  but  submerged  only 
twenty-five,  which  makes  it  an 
easy  load  for  a  single  diver. 
The  water-tight  covers  of  the 
camera  and  lens  compartments 

FIG'7'       nd™hngf°rWeS'        are  ga^keted  and  held  fast  by 

quick-acting  latches  which  re- 
quire no  tools  to  operate  and  make  loading  an  extremely  rapid  opera- 
tion. All  controls  are  clearly  calibrated  and  visible  from  the  operat- 
ing position.  The  diver-cameraman  can  accomplish  all  adjustments 
under  water  with  the  exception,  of  course,  of  reloading. 

We  have  also  a  housing  for  the  Bell  &  Howell  Eyemo,  Fig.  5.  In 
this  case  the  camera  is  removable  from  the  water-tight  case.  The  only 
permanent  alteration  is  the  addition  of  fittings  to  the  lenses  which 
in  no  way  interfere  with  the  ordinary  use  of  the  camera  in  air.  The 
case  will  accommodate  any  of  the  standard  Eyemo  lenses  up  to  the 
3  3/4-inch  focal  length.  Lenses  can  not,  however,  be  changed  under 
water.  Provision  is  made  for  the  underwater  adjustment  of  lens 
focus  and  aperture,  for  winding  of  the  spring  motor,  and  operation  of 
the  trigger.  All  controls  are  visible  and  adjustable  from  the  rear  of 
the  camera.  There  is  a  single  large  case  opening  which  is  gasketed 
and  held  by  quick-acting  latches,  and  it  is  not  necessary  to  remove 


Jan.,  1939] 


UNDERSEA  CINEMATOGRAPHY 


15 


the  camera  from  the  housing  which  reduces  to  a  minimum  the  time 
required  to  load  or  change  lenses  and  niters. 

There  is  also  available  a  professional  model  still  camera,  Fig.  6, 
which,  because  it  uses  standard  35-mm.  motion  picture  film,  can  be 


FIG.  8.     Underwater  range-finder. 

used  for  taking  check  stills  of  motion  picture  scenes  as  well  as  for  or- 
dinary still  pictures.  This  instrument  makes  use  of  the  Leica  camera 
mechanism,  and  all  controls  can  be  operated  by  the  diver  more  con- 
veniently than  in  the  average  air 
camera.  It  is  necessary  to  take  it 
to  the  surface  only  for  film  reloading. 
It  weighs  in  air  approximately  twenty- 
four  pounds  and  submerged,  about 
ten.  There  is  an  underwater  choice 
of  no  filter,  either  one  of  two  color 
filters,  a  polarizing  plate,  or  combina- 
tion of  filter  and  polarizing  plate. 
There  is  also  provided  an  indicator 
for  determining  the  proper  degree  of 
rotation  for  the  polarizing  plate.  A 
large  brilliant-image  field-finder  is  in- 
corporated in  the  camera  as  well  as  a 
built-in  range-finder  coupled  to  the 
camera  lens  in  the.  interest  of  rapid 
focusing. 

"For  determining  the  correct  exposure 
under  water,  which  is  essential  as  the 
light  available  varies  greatly  both  in 
amount  and  color  with, differences  in  depth,  the  amount  and  kind  of 
impurities  present,  arid  the  condition  of  the  water  surface,  there,  is  a 


PIG.  9.     Underwater  lamp. 


16  E.  R.  F.  JOHNSON  [j.  s.  M.  P.  E. 

substantial  water-tight  housing  for  a  Weston  model  650  exposure- 
meter,  Fig.  7. 

Accurate  measurement  of  distance  is  also  necessary.  Under  most 
conditions  illumination  is  relatively  weak  and  aperture  settings  must 
be  large,  resulting  in  short  focal  depth.  Under  water  the  eye  is  a  very 
poor  judge  of  distance  and  yardstick  measurements  are  often  difficult. 
To  make  quick  accurate  measurements  possible  an  underwater  range- 
finder  has  been  developed,  Fig.  8.  It  is  of  the  double-image  type, 
different  from  the  conventional  instrument  only  in  the  increased 
size  of  the  optical  parts  and  in  the  calibration  which  gives  true  dis- 
tances under  water. 


FIG.  10.    Amateur  underwater  still  camera. 

For  supplementing  natural  light  there  are  special  underwater 
lamps  (Fig.  9)  using  corrosion-resistant  reflectors,  which  for  ease  of 
handling  in  underwater  currents  are  of  comparatively  small  size. 
Special  diving  lamps  as  well  as  standard  photofiood  lamps  in  water- 
proof sockets  can  be  used  in  these  reflectors. 

Because  of  an  increasing  interest  on  the  part  of  amateurs  in  this 
field  we  have  designed  a  compact  still  camera  (Fig.  10).  It  takes 
pictures  2x/4  X  2x/4  inches  and  is  equipped  with  a  rapid  lens.  All 
necessary  controls  can  be  operated  under  water  and  filters  and  polar- 
izing plate  may  be  used.  It  incorporates  a  fitting  for  flash  bulbs 
which  is  synchronized  with  the  shutter.  The  only  concession  to  the 
low  price  demanded  of  an  amateur  camera  is  a  slight  sacrifice  in  the 
convenience  and  rapidity  of  operation.  In  a  short  time  we  expect  to 
have  a  similar  case  for  one  or  more  of  the  popular  16-mm.  motion 
picture  cameras. 


Jan.,  1939]  UNDERSEA  CINEMATOGRAPHY  17 

REFERENCES 

1  HULBURT,  E.  O.:    "On  the  Penetration  of  Daylight  into  the  Sea,"  J.  Opt. 
Soc.  Amer.,  XXII  (July,  1932),  No.  7,  p.  408. 

2  DARBY,  H.  H.,  JOHNSON,  E.  R.  F.,  AND  BARNES,  G.  W.:    "Studies  on  the 
Absorption  and  Scattering  of  Solar  Radiation  by  the  Sea,"   Carnegie  Inst.  of 
Washington,  Publication  No.  475  (Oct.  15,  1937),  p.  191. 

3  RAMAN:    "Molecular  Diffraction  of  Light,"  Proc.  Royal  Soc.,  101   (1922), 
p.  64. 

DISCUSSION 

MR.  KELLOGG:  For  getting  a  light  background,  do  you  ever  inject  into  the 
water  something  that  will  produce  a  milky  cloud  behind  the  subject  so  that  it  will 
not  obscure  the  picture? 

MR.  JOHNSON  :  We  have  never  attempted  to  do  that.  There  is  so  much  milki- 
ness  in  the  water  naturally  that  we  have  never  thought  about  adding  any.  The 
background  tends  to  come  up  and  hit  you  in  the  face.  You  are  seldom  more  than 
twenty  feet  away  when  this  nuisance  light  or  haze  light  obscures  everything. 
Generally  we  try  to  get  as  much  animal  or  plant  life  or  coral  growth  or  sand  bank 
or  other  physical  object  in  the  background  rather  than  to  leave  a  lot  of  this  fog. 

MR.  CRABTREE:  Have  you  used  artificial  light? 

MR.  JOHNSON:  Yes.  The  trouble  with  artificial  light  is  that  we  have  to  have 
such  a  tremendous  quantity  of  it  to  amount  to  anything.  The  red  rays  become 
dissipated  in  heat,  and  of  course  most  artificial  light  has  plenty  of  red  rays  in  it. 

MR.  CRABTREE:  Possibly  the  high-intensity  mercury- vapor  lamp  would  be 
useful.  It  would  at  least  be  water-cooled. 

MR.  JOHNSON:  It  would  be  a  better  lamp  than  some  of  the  others.  In  tank 
work  I  am  inclined  to  use  a  lot  of  artificial  light,  but  from  below  rather  than  up 
above. 

The  light  from  a  sodium  lamp  would  introduce  far  less  haze  than  that  from  a 
mercury  lamp,  but  for  underwater  photographic  use  a  lamp  of  high  efficiency 
over  a  wide  range  of  the  spectrum  should  be  selected.  We  therefore  do  not 
recommend  either  mercury-vapor  or  sodium  lamps  as  especially  useful  to  the 
underwater  cinematographer. 

MR.  CRABTREE:  I  noticed  a  tripod  in  one  of  the  pictures.  The  construction 
seemed  to  be  quite  different  from  that  of  the  normal  tripod. 

MR.  JOHNSON:  The  tripod  was  a  specially  built  brass  tripod,  made  like  a  slide 
trombone,  so  you  could  move  it  in  and  out  from  the  standing  position.  It  is 
made  very  solid,  because  under  water  one  is  pushed  around  a  lot  so  he  can  not  use 
a  surface  tripod. 


THE  ROAD  AHEAD  FOR  TELEVISION* 
I.  J.  KAAR** 

Summary. — Now  that  television  standards  have  been  agreed  upon  in  the  United 
States,  commercial  receiving  sets  will  undoubtedly  be  available  very  soon,  and  regularly 
scheduled  television  programs  may  be  expected  at  the  same  time.  How  good  will  the 
television  be  and  what  are  the  problems  yet  to  be  solved  before  television  reaches  the 
technical  maturity  that  radio  has  today?  These  are  questions  of  considerable  interest 
to  engineers  in  related  fields,  and  are  the  subject  matter  of  the  present  paper.  The 
quality  of  present-day  television  pictures  is  compared  with  that  of  motion  pictures 
both  in  the  theater  and  in  the  home.  A  discussion  is  given  of  the  problems  that  have 
been  solved  to  make  television  what  it  is  today,  and  consideration  is  given  to  the 
problems  that  must  be  solved  to  make  television  what  we  hope  it  will  be  tomorrow. 
The  problems  of  signal  propagation  and  interference  are  discussed,  and  the  matter 
of  network  program  distribution  is  considered.  Finally,  a  short  introduction  is 
given  to  the  commercial  problems  in  television. 

For  several  years  the  public  has  been  increasingly  curious  to  know 
when  television  would  be  introduced  commercially,  and  a  great 
variety  of  explanations  have  been  advanced  by  uninformed  persons 
as  to  why  it  has  not  happened  already.  Of  course,  at  first  the  reason 
was  lack  of  technical  quality;  but  in  the  past  few  years  the  quality 
of  pictures  achieved  has  certainly  been  good  enough  to  interest  an 
increasingly  large  proportion  of  the  population.  However,  two  major 
questions  had  still  to  be  answered  before  the  widespread  commercial 
introduction  of  television.  The  first  of  these  was  the  fixing  of  satis- 
factory television  standards  and  the  second  was  the  discovery  of  a 
satisfactory  method  of  paying  for  the  programs.  The  first  matter 
has  practically  been  settled;  the  second  has  not. 

Television  differs  from  sound  broadcasting  very  markedly  in  the 
importance  of  standards.  In  sound  broadcasting,  if  the  method  of 
modulation  (amplitude,  frequency,  or  phase)  is  once  determined,  any 
receiver  which  can  be  tuned  to  the  carrier  frequency  of  a  given  trans- 
mitter can  receive  its  program.  The  technical  quality  of  transmitted 

*  Presented  at  the  1938  Fall  Meeting  at  Detroit,  Mich. ;     received  October  21, 
1938. 

**  General  Electric  Company,  Bridgeport,  Conn. 

18 


THE  ROAD  AHEAD  FOR  TELEVISION  19 

programs  can  be  improved  year  by  year,  but  while  this  happens,  a 
receiver  once  purchased  is  always  usable,  even  though  it  may  become 
outmoded  as  compared  with  current  models.  The  situation  in  tele- 
vision is  quite  different.  Due  to  the  use  of  scanning  and  the  necessity 
of  synchronization  between  the  receiver  and  transmitter,  if  trans- 
mission standards  are  changed,  receivers  designed  for  the  old  stand- 
ards become  useless.  Because  of  this  fact,  no  responsible  manufac- 
turer would  sell  receivers  to  the  public  until  standards  were  fixed  by 


FIG.  1.     Typical  scene  in  British  television  studio. 

the  industry  and  sponsored  by  the  Federal  Communications  Com- 
mission. Furthermore,  American  manufacturers  did  not  desire  to 
fix  standards,  except  at  such  a  high  quality  that  widespread  and 
sustained  interest  on  the  part  of  the  public  would  be  assured  and  so 
that  adequate  provision  for  continued  perfection  was  possible.  It 
required  considerable  technical  perfection  to  justify  these  high  stand- 
ards, but  this  has  now  been  attained  and  the  essential  standards  have 
been  agreed  upon.  Consequently,  it  may  be  said  with  some  assurance 
that  the  last  technical  obstacle  in  the  path  of  commercial  television 
has  been  removed,  at  least  so  far  as  the  excellence  of  the  picture  under 
proper  conditions  is  concerned. 


20 


I.  J.  KAAR 


[J.  S.  M.  P.  E. 


The  question  of  who  shall  pay  for  television  programs  has  not  yet 
been  answered.  As  is  well  known,  the  cost  of  sound  broadcasting  is 
borne  by  "sponsors,"  who  pay  enough  for  their  own  programs  to 
enable  the  stations  and  networks  to  fill-in  the  unsponsored  time  with 
sustaining  programs  of  good  quality  and  to  make  a  profit  in  addition. 
However,  this  situation  now  requires  the  existence  of  tens  of  millions 
of  receivers  in  the  country  with  listeners  who  may  be  induced  to  buy 
the  advertised  products.  Such  an  audience  does  not  exist  in  television 


FIG.  2.     Typical  scene  in  American  television  studio. 

and  can  not  be  expected  for  several  years.  Of  course,  no  such  audience 
existed  in  the  early  days  of  sound  broadcasting  either,  and  the  re- 
ceiver manufacturers  themselves,  along  with  a  few  individual  com- 
panies who  built  stations  for  their  own  advertising  purposes,  operated 
the  stations.  In  those  days,  however,  the  thought  of  something  com- 
ing through  the  air,  receivable  at  no  cost,  was  an  entirely  new  one. 
People  were  quite  satisfied  with  the  new  toy  as  such  and  program 
excellence  was  a  secondary  consideration.  This,  of  course,  meant 
that  the  cost  of  broadcasting  (as  compared  to  the  present)  was  low. 
Now  the  public  has  been  educated  to  expect  a  high  degree  of  excellence 


Jan.,  1939] 


THE  ROAD  AHEAD  FOR  TELEVISION 


21 


in  program  material  and  it  is  doubtful  if  mediocre  program  material 
in  television  would  be  acceptable.  This  has  been  quite  strikingly 
proved  in  England.  In  other  words,  when  television  is  born,  it 
must  be  born  full-fledged  as  far  as  program  material  is  concerned. 
This,  of  course,  means  great  expense  which,  undoubtedly,  will  have  to 
be  borne  by  the  pioneers. 

In  Great  Britain  commercial  television  is  already  a  reality  and  it  is 
of  interest  to  consider  some  of  its  various  aspects.    American  tele- 


FIG.  3.     An  outside  pick-up  in  England. 


vision  will  be  quite  similar,  except  for  improvements  based  upon  the 
progress  of  the  art  since  the  British  standards  were  set. 

Fig.  1  is  a  photograph  of  a  television  studio  showing  the  general 
layout.  In  particular  there  is  seen  the  performer  and  the  position  of 
the  camera  tube  and  the  microphone. 

Fig.  2  is  a  similar  set-up  in  an  American  studio. 

Fig.  3  is  a  scene  showing  a  camera  tube  being  used  for  outside  pick-up 
in  England. 

Fig.  4  is  an  unretouched  photograph  of  an  image  on  the  screen  of 
a  picture  tube  in  England. 

Fig.  5  is  a  similar  picture  taken  in  America. 


22 


I.  J.  KAAR 


[J.  S.  M.  P.  E. 


FIG.  4.     Photograph  of  picture  tube  image  in  England. 


FIG.  5.     Photograph  of  picture  tube  image  in  America. 


Jan.,  1939]  THE  ROAD  AHEAD  FOR  TELEVISION  23 

Fig.  6  is  a  view  of  the  antenna  tower  of  the  London  (Alexandra 
Palace)  Television  Transmitter.  The  mast  carries  two  separate 
aerials,  vision  (above)  and  the  accompanying  sound  (below) . 

Fig.  7  is  a  view  of  the  interior  of  a  mobile  television  control  room 
in  England.  At  the  center  of  the  photograph  are  the  picture  monitors. 
This  equipment  is  mounted  in  a  "van"  and  has  been  used  very  suc- 
cessfully at  sporting  events.  The  signals  in  this  case  are  transmitted 
to  the  main  transmitter  at  very  short  wavelengths  and  rebroadcast 
at  high  power. 

Fig.  8  is  a  view  inside  an  American  "van"  serving  the  same  purpose. 

STANDARDS 

Let  us  next  briefly  consider  the  television  standards  which  have 
been  adopted  in  this  country  and  the  reasons  for  their  adoption.  The 
reader  is  no  doubt  acquainted  with  the  general  scheme  of  television 
used,  but  a  quick  review  of  the  essentials  may  be  in  order.  At  both 
the  camera  tube  and  the  picture  tube,  the  picture  is  scanned  by  an 
electronic  spot  (beam  of  electrons)  in  a  series  of  adjacent  horizontal 
lines.  The  number  of  these  lines  into  which  the  picture  is  divided 
in  the  scanning  process  determines  the  fineness  of  vertical  detail 
which  is  reproducible.  After  scanning  the  whole  picture,  the  elec- 
tronic spot  then  repeats  the  process  at  a  sufficiently  rapid  rate  so  that 
no  apparent  flicker  exists.  This  process  is  essentially  the  same,  so 
far  as  the  effect  upon  the  eye  is  concerned,  as  that  performed  by  the 
shutter  on  a  motion  picture  projector.  The  frequency  of  repetition 
of  scanning  of  the  whole  picture  is  known  as  the  frame  frequency. 

In  order  to  conserve  ether  space,  it  is  desirable  to  keep  the  frame 
frequency  as  low  as  possible.  Consequently,  an  artifice  is  employed 
in  order  to  increase  the  apparent  frequency  of  repetition.  This 
device  is  known  as  "interlace."  In  an  "interlaced"  picture  every 
other  line  of  a  picture  is  scanned,  and  after  the  whole  picture  has  been 
scanned  in  this  way,  the  lines  in  between  are  scanned.  This  gives 
the  physiological  effect  of  scanning  the  picture  twice,  as  far  as  flicker 
is  concerned,  even  though  all  details  of  the  picture  have  been  com- 
pletely scanned  only  once.  The  apparent  flicker  frequency  under 
these  conditions  which  is  twice  the  frame  frequency,  is  known  as  the 
field  frequency.  Now  obviously,  if  anything  other  than  a  complete 
blur  is  to  be  obtained,  it  is  necessary  that  the  number  of  lines  per 
frame,  the  order  of  scanning  of  the  lines,  and  the  number  of  frame_s 


24 


I.  J.  KAAR 


[J.  S.  M.  P.  E. 


per  second  be  identical  at  the  receiver  and  transmitter.    These,  ac- 
cordingly, have  been  standardized  in  America  as  follows : 

Number  of  lines  per  frame  =  N  =  441 
Number  of  frames  per  second  =  F  =  30 
Number  of  fields  per  second  —  60  (interlaced) 

To  these  we  may  also  add  the  standard  picture  aspect  ratio,  which 
is  4:3 — in  agreement  with  the  value  used  in  motion  pictures. 


FIG.  6. 


The  Alexandra  Palace  television  station  in 
London. 


There  is  a  reason  for  choosing  the  number  441  rather  than  some 
other  number  of  about  the  same  value.  It  may  be  shown  that  a 
necessary  requirement  for  a  stable  relationship  between  the  horizontal 
and  vertical  scanning  oscillators,  is  that  the  number  of  lines  per  frame 
be  a  whole  number  having  only  small  odd  factors.  If  no  factors 
larger  than  7  be  used,  Table  I  shows  the  list  of  possible  values  of  N. 
Four  hundred  and  five  lines  per  frame  is  the  figure  chosen  as  standard 
in  Great  Britain,  while  in  some  very  fine  laboratory  pictures  shown 
in  Holland,  567  lines  were  used.  


Jan.,  1939] 


THE  ROAD  AHEAD  FOR  TELEVISION 


25 


<to        O 

S      & 

•§    h 


I 


co 
">X 


t> 
«>  i>  X 


t-  10  X 
XioXXX*oXK>XXi>>oXX 
XcoXXXcoXoXXcoioXX 
XcoXXXcoXcoXXcocoXX 


coX 

CO 


coX       cocoX 

CO  CO 


00 


co 


XXcoXXcoXXcoXoXcotoX 
co  coX       co        coXcocoX 

CO  CO  CO 


CO    CO    CO    ^    T^    10 


CO 

co        10       i>  10  X 
-icoiOt-XXXXcoXcoXcoioX 

co        co        co  co  X 

co 


26  I.  J.  KAAR  [j.  s.  M.  P.  E. 

There  is  also  a  good  reason  for  using  30  as  the  frame  frequency.  It 
is  found  that  unless  the  frame  frequency  is  a  multiple  or  a  sub- 
multiple  of  the  power  supply  frequency,  a  shadow  will  move  across 
the  picture.  This  moving  shadow  has  about  the  same  physiological 
effect  as  flicker  and  is  very  disturbing.  However,  if  the  frame  fre- 
quency is  a  multiple  or  sub-multiple  of  the  power  line  frequency  the 
pattern  of  the  ripple  is  stationary  on  the  image  and  it  is  much  less 
objectionable.  Therefore,  since  60  cycles  is  standard  in  American 
power  distribution  systems,  30  frames  per  second  has  been  chosen  as 
standard  for  the  frame  frequency;  since  this  is  the  smallest  sub- 
multiple  of  60  whose  double  is  above  the  maximum  flicker  frequency 
observable  by  the  eye. 

Among  other  matters  requiring  standardization  are  the  synchroniz- 
ing operations  at  both  the  transmitter  and  receiver.  It  is  clear  that 
scanning  at  the  transmitter  and  receiver  must  be  exactly  synchronous 
to  within  an  extremely  small  error.  In  order  to  accomplish  this, 
synchronizing  signals  are  always  transmitted  with  the  picture  signals. 
The  purpose  of  these  synchronizing  signals  is  to  start  the  scanning 
of  both  the  lines  and  frames  at  exactly  the  right  time.  A  detailed 
investigation  of  synchronizing  signals  would  be  out  of  place  here, 
but  it  may  be  stated  as  absolutely  essential  that  the  type  of  synchron- 
izing signal  transmitted  should  be  completely  standardized. 

The  decision  as  to  which  was  the  most  desirable  synchronizing 
signal  was  one  of  the  most  difficult  of  all  questions  which  confronted 
the  Television  Standardizing  Committee.  The  signal  shown  in  Fig. 
9  was  ultimately  fixed  as  standard.  This  synchronizing  signal  is 
described  as  of  the  "serrated  vertical"  type.  It  is  believed  that  with 
the  use  of  this  signal,  the  most  stable  and  accurate  synchronization 
can  be  obtained.  Furthermore,  considerable  latitude  is  offered  to 
the  designer  as  to  the  means  chosen  for  utilizing  the  signal. 

The  next  subject  which  we  wish  to  consider  is  the  frequency  chan- 
nel width  required  in  television.  It  may  be  shown1  that  in  order  to 
transmit  the  available  intelligence  in  a  television  picture  with  N 
scanning  lines  per  frame,  and  a  frame  frequency  of  F,  a  minimum  fre- 
quency range  from  zero  to 

2  m 

is  required.    In  this  equation  R  is  the  aspect  ratio.    Substituting  into 
equation  1  the  values  which  have  been  standardized,  we  get 


Jan.,  1939]  THE  ROAD  AHEAD  FOR  TELEVISION 


27 


28  I.  J.  KAAR  [j.  s.  M.  P.  E. 


Thus  for  effective  utilization  of  the  intelligence  available  from  a 
standard  television  picture,  there  must  be  complete  and  undistorted 
transmission  of  all  frequencies  from  zero  to  at  least  2,750,000  cycles. 
If  this  signal  is  used  to  modulate  a  radio  frequency  carrier,  an  ex- 
tremely wide  frequency  channel  is  obviously  required. 

In  order  to  economize  on  the  use  of  the  frequency  band  thus  re- 
quired, single  side-band  transmission  is  proposed.  The  system  may 
more  properly  be  termed  "sesqui-side-band."  In  this  system,  the 
elimination  of  one  side-band  is  achieved  by  the  use  of  band-pass 
niters  which  have  a  range  of  partial  transmission  in  the  region  on 
either  side  of  the  transmission  band.  The  carrier  may  be  placed  in 
one  of  these  edge  bands  at  a  point  where  there  is  approximately  50 
per  cent  transmission.  It  may  be  shown  that  such  a  system  has  essen- 
tially double  side-band  transmission  for  very  low  frequencies,  and 
single  side-band  transmission  for  medium  and  high  frequencies.  To 
return  now  to  the  question  of  utilization  of  the  frequency  channel, 
it  is  noted  that  by  means  of  "sesqui-side-band"  transmission  the  fre- 
quency band  required  by  the  picture  signal  is  reduced  by  almost  50 
per  cent. 

In  transmitting  television  programs,  it  has  been  found  desirable 
to  transmit  the  picture  and  sound  in  the  same  channel.  This  allows 
a  single  oscillator  to  be  used  for  both  sight  and  sound  in  a  superhetero- 
dyne television  receiver,  thus  greatly  simplifying  tuning.  In  this 
system,  the  sound  and  sight  signals  are  separated  by  selective  circuits 
in  the  intermediate  frequency  amplifiers.  Fig.  10  is  a  diagram  of  a 
typical  television  receiver,  showing  how  it  transmits  and  separates 
the  picture  and  audio  signals. 

In  order  to  design  television  receivers,  it  is  necessary  that  the  rela- 
tive positions  of  the  audio  and  picture  signals  be  accurately  known. 
In  order  that  this  should  be  possible,  the  following  standards  have 
been  set  : 

Television  Channel  Width.  —  The  standard  television  channel  shall  not  be  less 
than  6  megacycles  in  width. 

Television  and  Sound  Carrier  Spacing.  —  It  shall  be  standard  to  separate  the 
sound  and  picture  carriers  by  approximately  4.5  MC.  This  standard  shall  go 
into  effect  just  as  soon  as  "single  side-band"  operation  at  the  transmitter  is  prac- 
ticable. (The  previous  standard  of  approximately  3.25  MC  shall  be  superseded.) 


Jan.,  1939] 


THE  ROAD  AHEAD  FOR  TELEVISION 


29 


<l 


CD! 


30 


I.  J.  KAAR 


[J.  S.  M.  P.  E. 


Sound  Carrier  and  Television  Carrier  Relation. — It  shall  be  standard  in  a  tele- 
vision channel  to  place  the  sound  carrier  at  a  higher  frequency  than  the  television 
carrier. 

Position  of  Sound  Carrier. — It  shall  be  standard  to  locate  the  sound  carrier  for  a 
television  channel  0.25  MC  lower  than  the  upper  frequency  limit  of  the  channel. 

In  addition  to  the  standards  already  mentioned,  there  are  certain 
other  standards  which  have  been  adopted,  and  these  will  be  com- 
mented upon  briefly.  Thus : 

It  shall  be  standard  in  television  transmission  that  black  shall  be  represented 
by  a  definite  carrier  level  independent  of  light  and  shade  in  the  picture. 

This  means  that  the  background  level  is  transmitted  in  a  television 
signal,  thereby  eliminating  the  need  for  readjustment  of  the  receiver 


FIG.  10.     A  typical  television  receiver. 


when  the  scene  being  televised  changes  from  a  preponderance  of 
white  to  a  preponderance  of  black. 

It  shall  be  standard  for  a  decrease  in  initial  light  intensity  to  cause  an  increase 
in  the  radiated  power. 

A  technical  description  of  this  standard  is  to  say  that  the  polarity  of 
the  transmission  is  negative.  It  is  seen  that  a  choice  exists  so  it  is 
necessary  that  this  point  be  standardized,  otherwise  the  picture  tube 
at  the  receiver  would  frequently  show  the  equivalent  of  a  photographic 
negative.  Even  more  important  than  this  is  the  fact  that  unless  the 
receiver  were  built  for  the  polarity  of  transmission  sent  out  by  the 
transmitter,  the  synchronizing  signals  would  not  be  effective. 


Jan.,  1939] 


THE  ROAD  AHEAD  FOR  TELEVISION 


31 


Percentage  of  Television  Signal  Devoted  to  Synchronization. — If  the  peak  ampli- 
tude of  the  radio  frequency  television  signal  is  taken  as  100  per  cent,  it  shall  be 
standard  to  use  not  less  than  20  nor  more  than  25  per  cent  of  the  total  amplitude 
for  synchronizing  pulses. 

Transmitter  Modulation  Capability. — If  the  peak  amplitude  of  the  radio-fre- 
quency television  signal  is  taken  as  100  per  cent,  it  shall  be  standard  for  the  signal 
amplitude  to  drop  to  25  per  cent  or  less  of  peak  amplitude  for  maximum  white. 

Transmitter  Output  Rating. — It  shall  be  standard,  in  order  to  correspond  as 
nearly  as  possible  to  equivalent  rating  of  sound  transmitters,  that  the  power  of 


TABLE  II 

Some  American  Picture  Tube  Characteristics 


Normal 

Operating 

Type 

Overall 
Tube 

Anode 
Voltage 

Spot 

De- 

Ify*P*e 

Diameter 

Length 

Size 

flec- 

Focus- 

(Inches) 

(Inches) 

(Volts) 

(Lines) 

tion 

ing 

Remarks 

3 

ny» 

1,500 

250 

5-5 

5 

Green  Screen 

White  Screen 

5V4 

157/8 

1,500-2,000 

375-425 

5-5 

5 

Green  Screen 

White  Screen 

5 

153/4 

3,000 

450 

M-M 

5 

Yellow-Green 

Screen 

White  Screen 

9 

21 

6,000 

450 

M-M 

5 

Yellow-Green 

Screen 

White  Screen 

12 

24V2 

6,000 

450 

M-M 

5 

White  Screen 

4"  Projection 

14V2 

20,000 

450 

M-M 

S-M 

Green    or    Yellow 

Green  Screen 

*  M-M  —  magnetic  deflection  both  ways. 

5-5    =  electrostatic  deflection  both  ways. 

**  5  =  electrostatic  focusing. 

S-M  =  combined  electrostatic  and  magnetic  focusing. 


television  picture  transmitters  be  nominally  rated  at  the  output  terminals  in 
peak  power  divided  by  four. 

Relative  Radiated  Power  for  Picture  and  for  Sound. — It  shall  be  standard  to 
have  the  radiated  power  for  the  picture  approximately  the  same  as  for  sound. 

The  last  four  standards  related  particularly  to  output  powers  and 
power  ratings,  and  while  they  are  important  in  regulating  the  design 
of  transmitters  and  receivers,  they  are  not  intimately  connected  with 
picture  quality,  which  is  of  principal  concern  here. 


32  I.  J.  KAAR  (j.  s.  M.  p.  E. 

THE  TELEVISION  PICTURE 

When  television  is  discussed  by  the  public,  the  questions  most  fre- 
quently asked  are  "How  good  is  television?"  "How  good  will  it 
be?"  and  "How  much  will  it  cost?"  The  answers  to  these  questions 
involve  such  matters  as :  How  large  will  the  picture  be  ?  How  bright 
will  it  be?  How  much  detail  will  it  show?  How  clear  will  it  be?  A 
discussion  of  these  considerations  will  be  of  interest. 

The  standard  high-quality  television  system  which  will  possibly 
be  commercialized  shortly  will  have  a  12-inch  tube  with  a  7!/2  by 
10-inch  picture.  Three,  5-,  7-,  and  9-inch  tubes  will  probably  also 
be  standard  commercial  sizes.  Compared  with  the  size  of  a  motion 
picture  or  even  a  home  movie,  these  dimensions  seem  small.  How- 
ever, considering  the  fact  that  the  audience  viewing  a  television 
picture  will  ordinarily  not  be  more  than  perhaps  four  feet  from  the 
screen,  and  in  the  case  of  the  small  tubes,  even  one  foot  from  the 
screen,  these  sizes  do  have  considerable  entertainment  value.  Any- 
one who  has  seen  good  pictures  on  9-inch  or  12-inch  tubes  will  testify 
that  when  the  program  is  interesting,  the  observer  forgets  that  he  is 
viewing  television  and  becomes  completely  absorbed  in  the  action 
on  the  screen.  Nevertheless,  it  is  reasonable  to  expect  larger  pictures 
in  the  best  systems  of  the  future.  Table  II  shows  the  characteristics 
of  some  present-day  television  tubes. 

The  matter  of  increasing  the  size  of  the  cathode-ray  picture  pre- 
sents some  serious  obstacles.  As  tubes  become  larger  they  also  be- 
come longer  and  their  overall  size  becomes  such  that  it  is  difficult  to 
find  suitable  cabinets  for  them,  which  at  the  same  time  lend  them- 
selves to  attractive  styling.  For  this  reason,  when  a  12-inch  tube 
is  used,  it  is  invariably  mounted  vertically  in  a  cabinet,  and  the  pic- 
ture is  seen  as  a  mirror  image  by  the  observer.  Since  a  mirror  causes 
a  loss  of  light,  and  possible  double  images  and  distortion,  it  is  an  un- 
desirable adjunct  at  best.  As  a  further  difficulty,  as  cathode  ray 
tubes  are  increased  in  size,  they  require  more  driving  power,  which 
is  expensive,  and  higher  anode  voltages,  which  besides  the  additional 
cost,  also  represents  a  shock  hazard.  Thus  the  prospect  of  making 
cathode-ray  tubes  for  home  use  with  screen  diameters  exceeding  12  or 
possibly  15  inches  does  not  seem  promising  at  this  tune. 

As  an  alternate  method  of  increasing  the  size  of  the  picture  obtain- 
able by  electronic  means,  the  projection  picture  tube  may  be  con- 
sidered. In  this  case  a  very  brilliant  picture  on  the  screen  of  a  4-inch 
cathode-ray  tube  is  enlarged  by  an  external  optical  system  and  is 


Jan.,  1939]  THE  ROAD  AHEAD  FOR  TELEVISION  33 

projected  on  a  screen  to  a  size  of,  say,  3X4  feet.  This  system  re- 
quires an  exceedingly  bright  tube  with  a  very  fine  spot.  The  ultimate 
size  of  projection  tube  pictures  is  limited,  on  the  one  hand,  by  the 
brightness  obtainable  from  a  fluorescent  screen  without  causing  its 
rapid  deterioration  and,  on  the  other  hand,  by  the  detail  which  can 
be  obtained  which  is  closely  associated  with  the  fineness  of  the  spot 
achievable.  Projection  tube  apparatus  is  probably  too  large,  compli- 
cated, and  costly  for  home  use,  but  for  public  performances  of  tele- 
vision programs,  it  undoubtedly  has  a  future. 

Mechanical  television  systems  have  also  been  used  for  obtaining 
large  pictures,  with  some  degree  of  success.  Of  these,  probably  the 
most  noteworthy  is  the  system  employed  by  Scophony.  This  system 
accomplishes  modulation  of  the  light-wave  by  utilizing  fringe  light, 
produced  by  virtue  of  passing  a  primary  beam  through  a  glass  vessel 
in  which  is  held  gasoline  or  benzine,  the  liquid  being  subjected  to 
vibration  from  a  quartz  crystal.  The  resulting  modulated  wave  is 
then  reflected  successively  by  two  rotating  mirrors  at  right  angles 
for  accomplishing  line  and  frame  scanning.  In  the  system  as  pro- 
posed, the  line  mirror  rotates  at  a  speed  somewhat  faster  than  30,000 
rpm. 

Closely  associated  with  the  problem  of  picture  size  is  the  problem 
of  picture  detail.  As  has  been  pointed  out,  the  vertical  detail  resolv- 
able in  a  picture  depends  upon  the  number  of  scanning  lines,  and  the 
horizontal  detail  depends  upon  the  ability  of  the  electrical  system  to 
pass  extremely  high  frequencies.  In  addition  to  this,  of  course, 
neither  can  go  beyond  the  effective  diameter  of  the  electron  spot. 
Observers  have  found  that  if  the  diameter  of  a  picture  element  sub- 
tends less  than  one  minute  of  arc  at  the  eye,  a  picture  contains  essen- 
tially all  the  detail  resolvable  by  the  observer.  If  the  observer  is  con- 
sidered to  be  4  feet  from  the  screen,  a  simple  calculation  will  show 
that  there  are  required  70  lines  per  inch  and  at  2  feet,  140  lines  per 
inch.  In  present-day  high-quality  pictures  on  a  12-inch  tube,  with  a 
71/z  inch  X  10-inch  picture,  and  400  useful  lines,*  there  are  53  lines 
to  the  inch.  It  is  not  unreasonable,  therefore  to  expect  the  number 
of  lines  in  television  pictures  to  be  a  matter  for  attention  in  the  years 
to  come. 

Goldsmith  states  that  a  high-quality  motion  picture  screen  has 
5,000,000  picture  elements.  This  would  be  equivalent  to  a  2000-line 

*  Ten  per  cent  of  the  441  lines  must  be  considered  lost  in  the  retrace  interval. 


34  I.  J.  KAAR  [j.  s.  M.  P.  E. 

picture,  which  would  give  1  degree  resolution  on  a  picture  3  feet  X  4 
feet  in  size,  viewed  from  a  point  5  feet  away.  While  it  is  not  too  much 
to  expect  such  television  pictures  sometime  in  the  future,  certainly 
a  great  many  problems  must  be  solved  first.  For  example,  such  a 
picture  would  require  150,000,000  picture  elements  per  second,  which, 
at  a  conservative  estimate,  would  need  a  band  width  of  80  megacycles 
per  program  for  its  transmission.  This  would  undoubtedly  require 
the  use  of  quasi-optical  carrier  frequencies  and  the  whole  problem 
would  entail  development  in  many  fields.  To  make  this  statement 
more  striking,  the  band  required  would  be  80  times  as  wide  as  the 
whole  spectrum  now  allocated  to  all  broadcasting  in  the  United  States. 

Another  important  consideration  in  television  development  is  the 
problem  of  picture  brightness.  Cathode-ray  tubes  used  in  television 
receivers  at  present,  are  as  bright  as  could  be  desired  in  a  darkened 
room.  Viewed  in  the  daylight,  however,  or  even  in  a  well-lighted 
living  room,  their  brightness  is  deficient.  While  it  is  always  possible 
to  darken  motion  picture  theaters,  television  receivers  will  probably 
be  expected  to  be  more  versatile,  and  to  operate  in  bright  light  as 
well. 

The  problem  of  increasing  picture  brightness  is  being  attacked  in 
many  ways.  Operating  voltages,  for  instance,  can  be  and  are  being 
increased.  This,  however,  is  undesirable  from  the  standpoint  of 
safety  and  cost.  More  efficient  luminescent  materials  are,  of  course, 
the  most  obvious  solution,  and  such  materials  are  constantly  under 
development. 

Another  interesting  development  in  this  connection  is  the  direct- 
viewing  tube.  This  differs  from  the  ordinary  tube  in  that  the  bom- 
barded side  of  the  screen  is  viewed,  instead  of  the  opposite  side,  as  is 
customary.  Such  tubes  naturally  require  a  construction  of  unortho- 
dox shape.  However,  they  may  be  the  tubes  of  the  future,  both  for 
reasons  of  brightness  and  also  for  reasons  of  contrast  and  detail,  as 
will  be  pointed  out  later.  Maloff 2  reports  a  direct-viewing  tube  having 
a  maximum  useful  brightness  of  100  candles  per  square-foot.  This 
is  more  than  ten  times  as  bright  as  the  highlights  in  a  high-quality 
motion  picture. 

Finally,  there  must  be  considered  the  matters  of  contrast  and  detail. 
The  present  contrast  available  in  television  tubes  is  quite  good,  but 
much  still  remains  to  be  done.  For  one  thing  a  cathode-ray  tube 
exhibits  the  phenomenon  of  halation.  This  is  the  optical  effect  of 
the  diffusion  of  light  in  the  screen  material,  and  with  it  we  may  also 


Jan.,  1939]  THE  ROAD  AHEAD  FOR  TELEVISION  35 

group  the  internal  reflection  of  light  from  the  walls  of  the  tube. 
Halation  is  well  known  in  photography.  It  decreases  the  brightness 
of  highlights  and  diffusely  lights  up  points  which  are  supposed  to  be 
dark,  particularly  in  locations  near  the  highlights.  The  general  effect 
is  thus  to  decrease  the  available  contrast  and  to  limit  the  possible  fine 
detail.  The  direct- viewing  tube  is  a  very  effective  means  of  decreas- 
ing halation.  When  such  a  tube  is  used,  the  increased  contrast  is 
very  striking. 

In  addition  to  halation,  a  cathode-ray  tube  also  exhibits  the 
phenomenon  of  "blooming,"  which  is  an  electrical  effect  and  results 
in  defocusing  the  spot  in  the  highlights.  Improved  focusing  arrange- 
ments can  be  used  to  decrease  "blooming,"  but  even  in  the  best  of 
modern  tubes  it  is  still  a  problem.  Since  the  contrast  desired  in  a 
television  picture  requires  an  electronic  beam  of  varying  density,  the 
focusing  of  the  tube  must  be  so  arranged  that  the  focal  point  does  not 
change  with  current  density,  i.e.,  brilliance.  This  is  not  an  easy  prob- 
lem. However,  it  is  evident  that  before  the  2000-line  pictures  men- 
tioned above  are  ever  obtained,  great  advances  must  be  made  in  the 
cure  of  "blooming." 

PROPAGATION 

The  problem  of  signal  propagation  in  television  assumes  an  im- 
portance which,  in  many  respects,  is  far  more  serious  than  that  of  the 
corresponding  problem  in  sound  transmission.  In  the  first  place,  the 
exceedingly  wide  frequency  channels  required  in  television  make  it 
necessary  that  the  signals  be  transmitted  in  the  ultra-short-wave 
bands.  At  these  frequencies,  as  is  well  known,  there  exists  reliably 
only  line-of -sight  transmission,  since  there  is  no  longer  reflection  from 
the  Heaviside  layer.  While  this  fact  limits  the  area  of  coverage  of 
any  transmitter,  it  is  actually  very  desirable  from  the  standpoint  of 
interference.  Thus  there  is  far  less  likelihood  of  multiple  images 
caused  by  multiple  path  reception,  due  to  reflections  from  the  Heavi- 
side layer,  or  of  interference  from  a  distant  station  operating  at  the 
same  frequency,  or  from  atmospheric  "static."  The  only  serious 
sources  of  noise  at  these  frequencies  are  those  generators  within  ap- 
proximately line-of -sight,  of  which  noteworthy  examples  are  automo- 
bile ignition  systems  and  medical  diathermy  machines. 

While  reflections  from  the  Heaviside  layer  are  negligible,  neverthe- 
less, because  of  the  very  short  waves  employed,  objects  such  as  steel 
buildings,  water  towers,  overhead  wires,  etc.,  provide  efficient  re- 
flectors and  give  rise  to  "ghost"  images.  The  severity  of  this  problem 


36  I.  J.  KAAR  [j.  s.  M.  P.  E. 

will  be  realized  much  more  fully  than  at  present  when  the  general 
public  begins  the  erection  of  receiving  antennae  and  the  operation 
of  receivers  on  a  large  scale. 

The  line-of  -sight  limitation  greatly  increases  the  difficulty  of  serv- 
ing a  large  geographical  area  with  a  given  program.  A  brief  con- 
sideration of  this  problem  will  be  of  interest.  It  can  logically  be  di- 
vided into  two  parts: 

(1)  The  conditions  necessary  for  adequate  coverage  of  the  line-of  -sight  area, 

and 

(2)  The  problems  involved  in  network  distribution. 

As  a  first  step  in  finding  the  conditions  necessary  for  adequate 
coverage  of  the  line-of  -sight  area,  we  recall  the  formula 

S  =  V2r  [Vh~i  +  VM  =  3560  (VJh  +  Vfh]  (3) 

where 

5  =  distance  over  which  line-of  -sight  transmission  takes  place  (in  meters) 
h\  =  height  above  intervening  ground  level  of  transmitting  antenna  (in  meters) 
h2  —  height  above  intervening  ground  level  of  receiving  antenna  (in  meters) 
r    =  the  radius  of  earth  (in  meters) 

This  formula  can  readily  be  derived  by  geometrical  consideration  of 
the  curvature  of  the  earth.  Next  consider  the  formula3  for  the  field- 
strength,  near  the  horizon,  from  a  transmitting  antenna  : 


E   =  voltspermeter  (4) 

Xr2 

where 

E  =  the  field  strength 

h     =  height  of  the  transmitting  antenna  (in  meters) 

a     =  height  of  above  effective  ground  of  the  receiving  antenna  (in  meters) 

T     =  distance  of  transmissions  (in  meters) 

X     =  wavelength  (in  meters) 

W  =  effective*  radiated  power  from  the  transmitter  (in  watts) 

Now  it  is  reasonable  to  expect  a  transmitting  antenna  to  be  located 
about  300  meters  above  ground  and  a  residential  receiving  antenna 
to  be  a  half-wave  dipole  located  4  meters  above  the  roof  (effective 
ground)  while  the  roof  itself  is  ten  meters  above  ground.  Under 
these  circumstances  there  results  from  equation  3: 

S  =  3560  (V300  +  Vl4)  =  75,000  meters 

=  75  km.  or  46.6  miles  (5) 

*  If  a  transmitting  antenna  other  than  a  half  -wave  dipole,  such  as  a  directional 
array,  is  used,  the  effective  value  of  W  may  be  increased  in  certain  directions. 


Jan.,  1939]  THE  ROAD  AHEAD  FOR  TELEVISION  37 

This  is  the  radius  of  the  area  over  which  reliable  coverage  can  be  ob- 
tained from  the  transmitter,  provided  that  the  power  of  the  trans- 
mitter is  sufficiently  great.  Consider,  now  what  this  transmitter 
power  must  be,  in  order  to  give  reliable  reception  at  the  distance  S 
from  the  transmitter. 

It  is  an  empirical  fact  that  reliable  reception  of  a  television  pro- 
gram requires  an  input  signal  of  about  one  millivolt.  Now  the  effec- 
tive height4  of  the  usual  half -wave  dipole  receiving  antenna  is  X/x. 
Therefore,  the  required  transmitter  antenna  power  is  given  by  the 
equation : 

88\/W  ah      \       in_, 
-J35-     '  -  =  10 

or 

W  =  1  28  X  lO-'    JP_  .=  1.28  X10-9  (75,000)* 
a2/*2  42  X  (300)2 

=  27,100  watts*  or  27.1  kw.  (6} 

Actually,  at  the  present  time  it  is  not  feasible  to  radiate  this  much 
power,  since  no  satisfactory  tubes  are  available  to  generate  it  at  these 
ultra  high  frequencies. 

Using  two  of  the  latest  high-power  developmental  tubes  in  push- 
pull,  it  is  possible  to  generate  10  kw.  (40  kw.  peak)  at  fifty  megacycles. 
The  limiting  factor  in  this  case  is  the  fact  that  the  size  of  high  power 
tubes  makes  it  impossible  to  tune  them  above  a  certain  critical  fre- 
quency and  their  high  interelectrode  capacities  make  it  difficult  to 
load  them  properly  and  still  preserve  the  desired  band-pass  character- 
istics. Thus  with  tubes  of  the  present  types,  it  is  not  yet  possible  to 
reach  the  desired  power  level;  and  the  condition  will  become  more 
serious  as  more  of  the  still  higher  frequency  channels  are  used  for 
television.  However,  it  is  reasonable  to  expect  that  the  ingenuity  of 
tube  designers  will  overcome  this  difficulty  in  the  next  few  years. 
In  the  meantime,  the  condition  can  still  be  corrected  by  increasing 
the  height  of  the  transmitting  antenna,  and  especially  of  the  receiving 
antenna. 

As  a  result  of  the  above,  an  interesting  fact  is  evident.  If  the  height 
of  the  receiving  antenna  is  neglected  in  calculating  the  line-of -sight 
distance,  there  results  : 

5  =  3560  vT 
*  Slightly  in  error  because  formula  4  is  extrapolated  beyond  the  horizon. 


38  I.  J.  KAAR  [J.  s.  M.  P.  E. 

Substituting  this  value  of  5  into  equation  6  there  results : 


It  is  evident  that  this  value  of  W  is  independent  of  h.  In  other  words, 
it  requires  12.9  kilowatts  of  transmitted  power  to  generate  a  signal 
of  one  millivolt  in  a  half -wave  dipole  4  meters  above  the  ground  at 
the  horizon.  This  value  is  independent  both  of  the  carrier  frequency 
and  of  the  height  of  the  transmitting  antenna.  The  latter  result  is 


FIG.  11. 


The  effect  of  multiple-path  transmission  or  reflection  upon  the 
received  image. 


very  surprising.  It  indicates  that  as  the  antenna  height  is  increased, 
the  same  power  still  suffices  to  reach  the  horizon — the  increased  dis- 
tance being  just  compensated  by  the  increased  antenna  height. 

Another  problem  of  considerable  importance  in  the  adequate  cover- 
age of  the  line-of -sight  area  is  the  elimination  of  multiple  reception 
or  echoes.  This  problem  is  of  practically  no  importance  in  sound 
broadcasting.  To  get  a  clear  idea  of  the  problem,  suppose  that  in 
addition  to  the  direct  ray  travelling  from  the  transmitting  to  the  re- 
ceiving antenna  there  is  also  a  ray  which  reaches  the  receiving  an- 
tenna by  way  of  reflection  from  a  large  building.  This  reflected  ray 


Jan.,  1939]  THE  ROAD  AHEAD  FOR  TELEVISION  39 

will  have  travelled  a  greater  distance  than  the  direct  ray  before  reach- 
ing the  receiver.  The  picture  which  it  carries  will  therefore  be  re- 
tarded in  time,  and  it  will  consequently  cause  a  similar  but  slightly 
displaced  picture  to  appear  next  to  the  desired  picture.  This  is  a 
very  annoying  effect,  and  great  effort  must  be  made  to  avoid  it.  This 
effect  is  illustrated  in  Fig.  II.5 

The  path  difference  necessary  to  cause  a  disturbing  echo  can  be 
easily  computed.  The  time  of  retardation  of  the  reflected  ray  is 
clearly  equal  to  the  difference  in  path  of  travel  divided  by  the  velocity 
of  light.  Then,  remembering  that  the  electron  beam  scans  (4/s  X 
30  X  441  X  441)  picture  elements  per  second,  the  displacement  of 
the  echo  from  the  main  picture  (in  picture  elements)  is 

D    =  4/3  X  30  X  441  X  441 

3  X  108 
=  0.026  times  the  path  difference  in  meters 

In  other  words,  a  path  difference  of  127  feet  will  cause  an  echo  dis- 
placement of  one  picture  element.  This  is  enough  to  detract  from 
the  quality  of  the  picture. 

The  elimination  or  reduction  of  echoes  is  a  complicated  problem. 
In  metropolitan  areas,  due  to  the  presence  of  many  reflectors  in  the 
form  of  tall  buildings,  the  problem  is  serious  indeed.  The  usual 
solution  is  to  use  a  directional  antenna  which  will  discriminate  against 
the  undesired  signal.  Horizontal  polarization  of  the  radiated  signal 
has  been  found  to  improve  the  signal-to-noise  ratio  at  television  car- 
rier frequencies,  and  its  use  will  therefore  probably  become  a  standard 
practice. 

Some  of  the  problems  connected  with  the  chain  distribution  of 
television  programs  may  now  be  considered.  There  are  two  general 
methods  which  have  been  used  to  transmit  television  programs  from 
a  key  transmitter  to  a  distant  transmitter.  These  are  the  use  of 
(1)  the  radio  relay  or  (2)  the  coaxial  (or  other)  high-fidelity  cable 
relay. 

Whichever  method  is  used,  the  relay  stations  must  be  sufficiently 
close  together  so  that  non-fading  noise-free  signals  are  received  at 
each  repeater  location.  It  has  been  found  that  relay  stations  must 
be  located  from  30  to  70  miles  apart,  the  exact  distance  depending  on 
noise  conditions  and  (in  the  case  of  the  radio  relay)  on  the  topography 
of  the  landscape.  It  has  been  customary  to  operate  radio  relays  at 
wavelengths  of  two  meters  or  less.  Each  relay  station,  of  whichever 


40  I.  J.  KAAR  [j.  s.  M.  P.  E. 

type,  must  reproduce  the  incoming  signal  with  the  highest  fidelity, 
having  neither  amplitude,  frequency,  nor  phase  distortion.  In  other 
words,  the  picture  must  not  be  degraded  in  passing  through  the  re- 
lays. 

It  is  not  surprising  that  the  great  problem  in  the  relaying  of  tele- 
vision signals  is  cost.  The  cost  per  mile  of  a  coaxial  cable  required  to 
handle  the  exceedingly  wide  frequency  bands  of  television  programs 
is,  at  the  present  time,  many  times  as  great  as  the  cost  of  correspond- 
ing networks  used  in  sound  broadcasting,  both  as  regards  initial  cost 
and  maintenance.  If  radio  relaying  is  used,  the  cost  of  the  relay  trans- 
mitters required  is  obviously  very  great.  However,  the  coming  years 
are  likely  to  bring  great  reductions  in  the  costs  of  both  methods  of 
relaying,  particularly  the  coaxial  cable. 

This  paper  has  been  an  effort  to  point  out  the  fact  that  many 
problems  still  must  be  solved  before  fully  satisfying  television  pic- 
tures will  be  available  in  the  home.  However,  it  is  not  to  be  con- 
strued that  the  commercial  introduction  of  television  will  await  a 
solution  to  these  problems.  Undoubtedly  television  will  be  com- 
mercialized in  the  near  future  and  the  problems  will  be  solved  as 
time  passes — much  the  same,  for  instance,  as  was  the  case  in  the  mo- 
tion picture  industry.  One  fact  is  very  clear,  that  the  further  de- 
velopment of  television  must  come  largely  through  findings  in  the 
field,  that  is,  by  actual  trial. 

The  author  wishes  to  acknowledge  the  assistance  of  Dr.  Stanford 
Goldman  in  the  preparation  of  this  paper  and  the  courtesy  of  the  Na- 
tional Broadcasting  Company  and  the  British  Broadcasting  Com- 
pany for  the  use  of  photographs  of  their  equipment. 

REFERENCES 

1  WHEELER,  H.  A.,  AND  LOUGHREN,  A.  V. :  Proc.  I.R.E.,  26  (May,  1938),  No.  5, 
p.  558. 

2  MALOFF,  I.  G.:  RCA  Review,  2  (Jan.,  1938),  No.  1,  p.  291. 

3  BEVERAGE,  H.  H.:   Monograph  "Television,"  RCA  Review,  2  (1938),  p.  99. 

4  HUND,  A.:      "Phenomena  in  High-Frequency  Systems,"  p.  466.     (McGraw- 
Hill.) 

6  SEELEY,  S.  W. :  "Effect  of  Receiving  Antenna  on  Television  Reception 
Fidelity,"  RCA  Review,  2  (April,  1938),  p.  435. 

DISCUSSION 

MR.  McNABB:  Referring  to  the  reproductions  (Figs.  4  and  5)  of  a  British  pic- 
ture and  an  American  picture,  the  line  structure  was  quite  evident  in  the  British 
picture,  but  the  contrast  seemed  a  little  better.  Is  the  contrast  better  in  the  Brit- 
ish picture  due  to  the  method  of  transmission,  or  is  the  transmission  of  direct 


Jan.,  1939]  THE  ROAD  AHEAD  FOR  TELEVISION  41 

current  along  with  the  signal  better  than  the  American  method  of  adding  the  d-c. 
at  the  receiving  end? 

MR.  KAAR:  There  is  no  essential  difference  in  the  method  of  transmission  in 
England  and  here.  The  only  difference  is  the  means  of  synchronization.  As  far 
as  contrast  and  detail  are  concerned,  there  should  be  no  difference  between  the  two 
systems  except  for  the  possible  fact  that  we  have  441  lines,  whereas  they  have  405. 

It  is  possible  to  photograph  any  kind  of  picture  from  the  front  of  a  picture  tube 
and  we  can  so  adjust  focus  and  contrast  as  to  make  the  line  structure  visible  on 
an  American  picture. 

As  a  matter  of  fact,  neither  of  these  pictures  is  a  good  example  because  they 
have  both  been  degraded  by  photographic  processes  in  the  original  photograph, 
the  enlargement,  the  negative,  and  the  lenses,  so  in  order  to  compare  the  two 
fairly  the  originals  should  actually  be  seen.  Our  pictures  are  somewhat  better 
than  the  British  pictures. 

MR.  FINN:  In  the  choice  of  repeaters,  Mr.  Kaar  suggested  that  the  choice  as 
between  coaxial  cables  and  straight  etherization  of  a  program  is  very  close.  Is  it 
your  suggestion  that  the  coaxial  cable  be  used,  over  hill  and  dale  for  thirty  or 
sixty  miles,  throughout  the  whole  broadcast  circuit,  to  blanket  the  country? 

MR.  KAAR:  That  is  a  difficult  question  to  answer  because  I  am  not  familiar 
with  the  recent  progress  on  coaxial  cable.  You  will  find  a  description  of  the  New 
York-Philadelphia  cable  in  the  literature.  As  I  remember,  it  has  repeaters  every 
ten  miles  and  as  yet  will  not  transmit  the  full  band  required.  Perhaps  some  day 
transcontinental  cables  may  be  laid  capable  of  handling  television  programs,  but 
I  can  not  say  that  they  will.  The  other  system  is  satisfactory  and  has  been  tried. 
As  to  the  economic  balance  between  the  future  use  of  cables  and  ether  channels, 
that  still  remains  to  be  answered. 

MR.  GOLDSMITH:  The  New  York-to-Philadelphia  cable  was  said  to  have  cost 
$540,000.  Whether  that  included  large  engineering  developmental  expenses  or 
not,  it  is  now  known.  In  any  case,  that  would  have  indicated  a  per-mile  cost  of 
$5000  or  $6000.  The  major  broadcasting  networks  in  the  United  States  today 
use  somewhere  on  the  order  of  40,000  or  45,000  miles  of  lines,  and  if  one  multiplies 
that  by  $5000  for  the  cost  of  laying  a  similar  coaxial  cable  network,  the  result 
of  the  multiplication  is  an  extremely  large  and  uneconomic  amount. 

However,  it  is  believed  likely  that  development  will  lead  ultimately  to  less 
costly  coaxial  cables  with  repeater  stations  closer  together  and  satisfactory  for  the 
purpose,  or  to  economic  radio  relay  systems  that  will  work  very  effectively. 

MR.  KAAR:  The  fact  that  such  a  serious  problem  exists  in  chain  programs  comes 
pretty  closely  home  to  the  motion  picture  engineer,  because  for  the  immediate 
present  there  is  an  answer  to  the  chain  broadcasting  of  television  programs, 
namely,  the  transmission  of  motion  picture  films,  which  will  undoubtedly  be  done 
extensively. 

MR.  GOLDSMITH:  There  are  many  practical  and  artistic  reasons  why  film  will 
necessarily  be  widely  used. 

MR.  WILLIFORD:  Does  the  adoption  of  the  60-cycle  frequency  as  standard 
mean  that  communities  having  25-  or  50-cycle  power  supply  are  definitely  out  of 
the  picture  as  far  as  television  is  concerned? 

MR.  KAAR:  There  is  no  connection  between  the  synchronizing  mechanism  of 
television  and  the  power  frequency.  The  synchronizing  is  accomplished  by 


42  I.  J.  KAAR  [j.  s.  M.  P.  E. 

transmitted  signals.  The  only  reason  for  and  the  advantage  of  choosing  a  frame 
frequency  that  is  a  multiple  or  submultiple  of  the  power-line  frequency  is  this:  If 
a  system  should  develop  a  ripple,  as  we  know  it  in  audio  work,  that  ripple  would 
occur  at  power-line  frequency.  If  the  frame  frequency  occurred  at  some  other 
frequency  than  that,  this  ripple,  which  would  be  either  a  light  area  or  a  dark  area, 
would  travel  across  the  screen.  If  the  system  is  perfect  and  there  is  no  ripple, 
it  makes  no  difference  at  all.  This  is  simply  chosen  as  a  safety  measure. 

MR.  GOLDSMITH:  If  the  power-supply  system  of  the  receiver  and  its  shielding 
are  so  engineered  that  no  such  effects  appear,  the  receiver  can  be  used  equally  well 
regardless  of  the  power  supply. 

MR.  CABLE:  It  seems  to  me  that  the  frequency  chosen  as  30  places  a  definite 
limitation  on  the  picture  brightness,  because  the  frequency  is  a  function  of  bright- 
ness. 

MR.  GOLDSMITH:  The  present  standard  is  60  pictures  per  second.  We  see  60 
"half  pictures,"  with  interlaced  scanning.  First  is  shown  a  picture  with  lines 
1,  3,  7,  and  so  on,  as  a  full  picture;  and  the  one  with  lines  2,  4,  6,  8,  and  so  on,  as 
the  next  picture,  a  sixtieth  of  a  second  later.  So  the  frame  frequency  is  30  but  the 
field  frequency  is  60  per  second.  You  substitute  for  picture  flicker  a  new  effect 
called  inter-line  flicker,  which  is  practically  invisible. 

MR.  FRIEDL:  In  selecting  the  number  of  frames  projected,  you  have  evidently 
regarded  power-supply  frequency  as  an  important  factor.  Inasmuch  as  the  mo- 
tion picture  film  will  be  used  as  a  means  of  widely  distributing  the  program,  the 
frame-frequency  of  the  motion  picture  is  a  consideration.  I  would  judge,  from 
the  decision,  that  the  more  difficult  matter  of  control  is  the  power  supply,  but  we 
as  motion  picture  engineers  naturally  ask  why  the  24-frame  frequency  with  in- 
terlacing to  give  48  images  was  not  considered  the  more  important  factor. 

You  speak  of  standards  in  television.  We  are  very  standards-conscious  in  the 
SMPE  and  are  aware  of  the  importance  of  international  as  well  as  national  stand- 
ardization. I  see  a  lack  of  uniformity  among  the  standards  adopted  by  Germany, 
Great  Britain,  France,  and  America.  That  might  be  excused  at  the  moment  be- 
cause of  the  fact  that  the  range  of  transmission  is  so  limited  and  we  can  not  expect 
immediately  to  transmit  across  the  ocean;  but  inasmuch  as  the  number  of  lines 
selected  is  441,  which  has  been  selected  mainly  to  allow  room  for  improvement,  can 
not  we  also  anticipate  improvement  and  have  confidence  in  the  effect  of  the  de- 
velopment to  look  forward  to  transmission  across  the  ocean,  and,  therefore,  inter- 
national standardization? 

MR.  GOLDSMITH:  We  may  hope  for  this,  because  some  such  standard  as  441 
lines  for  the  picture  might  be  adopted  by  all  the  nations.  But  it  must  be  ad- 
mitted that  at  the  present  time  radio  differs  from  motion  pictures  in  that  interna- 
tional standardization  is  rather  conspicuously  absent.  However,  it  may  come 
with  television. 

MR.  FRIEDL:  We  are  conscious  of  the  high  voltages  in  the  larger  tubes — 25,000 
and  40,000  volts.  What  is  the  voltage  on  the  12-inch  tube  and  how  does  the  sys- 
tem meet  with  the  protective  requirements  of  the  NFPA  and  the  Fire  Under- 
writers? 

MR.  KAAR  :  The  voltage  on  the  12-inch  tube  will  probably  be  6000  volts.  That 
sounds  like  a  very  serious  matter,  but  really  it  is  not.  If  you  sit  in  a  dentist's 
chair  and  he  turns  the  X-ray  on  you,  that  is  about  40,000  volts.  It  is  protected. 


Jan.,  1939]  THE  ROAD  AHEAD  FOR  TELEVISION  43 

It  simply  means  we  have  a  job  of  protecting  the  television  receiver,  possibly  by  an 
interlocked  back. 

MR.  FRIEDL:  All  I  can  say  is  that  conditions  in  the  home  where  children  might 
come  in  contact  with  the  apparatus  are  different  from  what  they  are  in  a  dentist's 
office. 

MR.  GOLDSMITH:  The  back  of  the  receiver  is  an  expanded  metal  mesh.  If  you 
open  the  back,  you  will  open  all  power  circuits  and  discharge  the  high-voltage 
condensers  automatically.  If  you  try  to  take  the  cathode-ray  tube  out  you  will 
similarly  open  up  the  circuits.  You  can  not  get  into  contact  with  a  high  voltage. 
It  is  generally  so  arranged  that  even  people  with  screw-drivers  and  determination 
simply  can  not  get  into  trouble,  and  we  hope  these  practices  will  continue. 

MR.  FRIEDL:  Does  horizontal  polarization  mean  that  the  antenna  will  be  hori- 
zontal? Also,  is  that  discussion  of  a  three-meter  receiving  antenna  going  back  to 
a  multiplicity  of  "wash  line"  antennas  on  every  roof? 

MR.  GOLDSMITH:  The  antenna  wire  or  rod  is  only  about  six  feet  long.  The 
two  component  rods  are  each  about  three  feeet  long. 

MR.  KAAR:  They  are  half  a  wavelength  long,  and  the  wavelengths  are  of  the 
order  of  five  meters. 

MR.  McNABB:  In  an  article  about  six  months  ago  in  Electronics,  regarding  the 
quality  of  television  pictures,  it  was  the  opinion  of  certain  American  engineers 
who  investigated  the  British  pictures  that  the  British  were  ahead  of  us  in  their 
technical  developments  as  well  as  their  commercial  exploitation  of  the  art. 
That  seems  to  disagree  with  the  opinions  of  other  American  engineers.  Exactly 
what  are  we  to  believe? 

MR.  GOLDSMITH  :  The  consensus  of  engineering  opinion  among  those  who  have 
seen  television  pictures  in  London  and  New  York  is  that  there  is  little  if  anything 
to  choose  between  them.  It  is  most  unlikely  that  practice  in  either  case  is  far 
ahead  of  the  other. 


REPORT  OF  THE  STUDIO  LIGHTING  COMMITTEE* 

Summary — In  a  previous  report  of  the  Studio  Lighting  Committee  the  need  of  a 
catalogue  of  studio  lighting  equipment  was  emphasized.  A  number  of  papers  have 
been  published  which  describe  various  lamps  and  light-sources  in  detail,  but  there 
has  not  been  assembled  in  one  paper  a  symposium  of  all  types  of  equipment  and  light- 
sources  used  on  motion  picture  sets. 

This  report  covers  all  types  of  equipment  in  general  use.  The  various  lighting  units 
are  numbered  and  briefly  described.  Photographs  of  popular  lamps  are  shown. 
Tables  give  minimum  and  maximum  beam  divergences,  carbon  and  bulb  sizes. 
Reference  numbers  are  assigned  to  the  various  lamps  for  convenience  in  listing  their 
characteristics.  Data  on  light  control  devices  and  lamp  filters  are  included. 

CARBON  ARC  LAMPS 

(1)  MR  Type  27  Scoop. — Chromium  plated  reflector  and  Factrolite 
glass  diffuser.    Solenoid  controlled.    A  twin-arc  flood  source,  used  for 
overhead  illumination  of  walls,  backings,  and  other  areas  that  can 
not  be  lighted  satisfactorily  by  spotlamps.    Suspended  singly  or  in 
groups.    A  smooth,  general-purpose  light-source. 

(2)  MR    Type   29   Broadside. — Chromium   plated   reflector   and 
Factrolite  glass  diffuser.    Solenoid  controlled.    A  twin-arc  flood  source 
that  may  be  raised,  lowered  and  tilted,  and  used  as  a  floor-lighting 
unit  for  building  up  front  light  to  the  desired  exposure  level. 

(3)  MR  Type  40  Duarc  Broadside. — Chromium  plated  reflector 
and  pebbled,  sand-blasted  Pyrex  glass  diffuser.    An  unproved  motor- 
controlled  twin-arc  flood-lamp  that  takes  the  place  of  both  scoop 
and  broadside  of  the  older  types. 

(4)  MR  Type  65  Arc  Spotlamp. — Eight-inch  diameter  Fresnel- 
type  lens.    High-intensity  rotating  mechanism.    Used  for  front  and 
back  lighting,  close-up  and  medium  shots.    The  intensity  is  almost 
uniform  in  the  main  portion  of  the  beam,  tapering  off  at  the  edges 
to  permit  overlapping  adjacent  beams  without  producing  objection- 
able high-intensity  zones.    Within  its  energy  capacity  this  lamp  may 
be  used  for  all  photographic  spot  lighting. 

(5)  MR  Type  90  Arc  Spotlamp. — Fourteen-inch  diameter  Fresnel- 
type  lens.     High-intensity  rotating   mechanism.     Used   for   back 

*  Presented  at  the  1938  Fall  Meeting  at  Detroit,  Mich. 
44 


STUDIO  LIGHTING  COMMITTEE  REPORT 


45 


FIG.  1.     Typical  high-intensity  rotating  element. 


UPPER   CARBON 
ACTUATING  LEVER 


INDUCTIVE  RES.  N0.2  INDUCTIVE  RES. NO.  I 

FIG.  2.     Typical  solenoid  feed  mechanism. 


46 


STUDIO  LIGHTING  COMMITTEE  REPORT     [j.  s.  M.  p.  E. 


Lamp  No. 

1.  MR  type  27  scoop. 

2.  MR  type  29  broadside. 

3.  MR  type  40  duarc  broadside. 


Jan.,  1939] 


STUDIO  LIGHTING  COMMITTEE  REPORT 


47 


lighting,  sunlight  effects  through  doorways  or  windows,  etc.,  for  key 
lighting  on  sets  of  moderate  size,  and  for  general  front  lighting  into 
the  rear  areas  of  deep  sets. 

(6)  MR  Type  170  Arc  Spotlamp.— Twenty-inch  diameter  Fresnel- 
type  lens.  High-intensity  rotating  mechanism.  Used  for  back,  cross, 
and  key  lighting,  and  for  wide-  and  narrow-angle  front  and  effect 


Lamp  No. 

4.  MR  type  65  arc  spotlamp. 

5.  MR  type  90  arc  spotlamp. 

6.  MR  type  170  arc  spotlamp. 
8.     36-inch  sun  arc. 


lighting.     This  unit  has  wider  use  for   both  black-and-white  and 
color  photography  than  any  of  the  other  arc  units. 

(7)  24-Inch  Sun  Arc. — Twenty-four  inch  diameter  glass  mirror. 
High-intensity  rotating  mechanism.  Normally  used  with  the  arc 
crater  facing  the  mirror  and  a  clear  glass  door  on  the  front  of  the 
lamp  house.  Where  very  sharp  shadows  are  necessary  the  clear 
glass  door  may  be  moved  to  the  position  normally  occupied  by  the 
mirror.  A  metal  door  is  then  placed  on  the  open  end.  A  large  number 


48  STUDIO  LIGHTING  COMMITTEE  REPORT     [J.  S.  M.  P.  E. 

of  these  lamps  have  been  converted  to  use  the  same  optical  system  as 
the  MR  type  1 70  lamp .  Used  for  back  lighting,  sunlight  effects  through 
windows  and  doorways,  etc.,  for  key  lighting  on  sets  of  moderate 
size,  and  for  general  front  lighting  into  the  rear  areas  of  deep  sets. 

(8)  36-Inch  Sun  Arc. — Thirty-six  inch  diameter  glass  mirror. 
High-intensity  rotating  mechanism.  Similar  to  the  24-inch  Sun  Arc 
except  as  to  size.  The  24-inch  Sun  Arc  is  rapidly  being  converted  to 
the  use  of  the  Fresnel  type  of  lens,  but  due  to  its  great  penetrating 
power,  the  36-inch  Sun  Arc  is  valuable  on  extremely  long  throws  and 
retains  its  popularity  in  its  present  form.  When  a  large  quantity  of 

diffused  light  is  required  from  this 
unit,  a  diverging  door  composed  of 
strips  of  cylindrical  lenses  replaces  the 
plain  glass  door.  The  lamp  is  used 
where  a  very  high  intensity  of  pro- 
jected light  is  required,  as  in  back 
lighting  behind  a  high  level  of  fore- 
ground illumination;  or  where  well 
denned  shadows  are  required ;  or  where 
a  clearly  defined  streak  of  light  is 
required  through  the  general  illumina- 
tion; or  for  producing  a  general  illu- 
mination of  great  penetration  and  high 
intensity. 

fl  ~~J*  ^    so~AmPere    Rotary    Arc    Spot- 

lamp. — An  8-inch  diameter  plano-con- 
Lamp  No.  10. 

vex  condenser  or  12 -inch  Fresnel-type 

lens.  High-intensity  rotating  mechanism.  One  of  the  early  high- 
intensity  arc  spotlamps.  This  lamp  is  not  suitable  for  color  in  its 
present  form  because  of  the  spectral  energy  distribution  of  the  carbon 
trim.  A  number  of  these  lamps  have  recently  been  converted  to 
the  use  of  11  mm.  X  20-inch  H.  I.  motion  picture  studio  positive 
carbons  to  make  them  suitable  for  color  photography.  Used  for 
back  lighting  on  black-and-white  sets  and  to  increase  the  intensity 
of  illumination  at  any  point  where  projected  light  is  required  within 
the  range  of  its  intensity. 

(10)  B  &  M  Type  9  Twin-Arc  Broadside.— Chromium  plated  re- 
flector and  glass  diffuser.  Solenoid  striker  with  direct  motor  feed. 
A  twin-arc  flood  source  that  may  be  raised,  lowered  and  tilted,  and 
used  as  a  floor-lighting  unit  for  use  in  building  up  front  light. 


Jan.,  1939]  STUDIO  LIGHTING  COMMITTEE  REPORT  49 


TABLE  I 

Arc  Lamps  for  Set  Lighting 

Ppsi-  Nega- 

*Degrees  Beam  tive  tive 

Lamp                                                                                              Divergencies  Carbon  Carbon 

No.                                       Unit                                             Min.               Max.  No.  No. 

1  MR  27  Scoop1                                  90            90  1  10 

2  MR  29  Broadside1                             90             90  1  10 

3  MR  40  Broadside                             90            90  1  10 

4  MR  65  Spotlamp*                               8             44  2  11 

5  MR  90  Spotlamp3                               8             44  5  14 

6  MR  170  Spotlamp2                             8             48  6  15 

7  24-Inch  Sun  Arc2                           **10             24  6  13 

8  36-Inch  Sun  Arc^                            10             32  6  13 

9  80- Amp.  Rotary  Spot*                     **8             30  4  12 
94             80- Amp.  Rotary  Spot 

(Converted)                                       8             44  3  12 
10                B  &  M  Type  9  Twin- 
Arc  Broadside                                90             90  1  10 


*  Approximate  figures  referring  to  usable  photographic  light. 
**  With  Fresnel-type  lens  divergences  are  approximately  8  to  44  degrees. 


TABLE  II 

Carbons  for  Set  Lighting 

Car- 
bon Arc 

No.  Positive  Carbons  Amperes  Volts 

1  8-Mm.  X  12  NP  MP  Studio2'"'7'"'"  38-43  35-40 

2  9-Mm.  X  20"  Hilow  Projector6'9  65-70  52-54 

3  11-Mm.  X  20"  HI  MP  Studio9  90-95  62-65 

4  1/2*  X  12"  80- Amp.  Rotary  Spot2'7'8'9  75-80  50-55 

5  13.6-Mm.  X  22"  HI  Projector4'5'9  110-115         54-56 

6  16-Mm.  X  20"  HI  MP  Studio2'4'5'7'8'9  140-150         64-67 


Negative  Carbons 

10  8-Mm.  X  12"  NP  MP  Studio 

11  7-Mm.  X  9"  Cored  Suprex  Negative 

12  Vs"  X  9"  Cored  80-Amp.  Rotary  Spot  Negative 

13  11-Mm.  X  10"  Plain-Cored  MP  Studio  Negative 

14  3/s"  X  9"  Cored  Orotip  Negative 

15  7/18"  X  9"  Cored  Orotip  Negative 


50 


STUDIO  LIGHTING  COMMITTEE  REPORT      [j.  s.  M.  P.  E. 


Lamp  No. 

20.  MR  type  36  studio  spotlamp. 

21.  MR  type  26  studio  spotlamp. 

22.  MR  type  16  cinelite. 

23.  MR  type  45  rifle  lamp. 

24.  MR  type  220  18-inch  sun  spot. 

25.  B  &  M  type  T5  studio  spotlamp. 


Jan.,  1939] 


STUDIO  LIGHTING  COMMITTEE  REPORT 


51 


INCANDESCENT  LAMPS 

(20)  MR  Type  36  Studio  Spotlamp—A  6-inch  diameter,  9-inch 
focus  plano-convex  condenser.    For  use  where  a  full  controlled  beam 
of  light  is  required:     in  close-up  photography  for  back  and  close 
lighting,  particularly  where  the  photography  demands  high  contrast 
of  light  and  shadows ;  in  general  motion  picture  set  lighting  it  is  use- 
ful for  special  effects  and  for  illuminating  areas  that  can  be  reached 
only  by  projected  light. 

(21)  MR  Type  26  Studio  Spotlamp. — A  spherical  mirror  is  adjust- 
ably mounted  behind  the  bulb  to  collect  the  rays  of  light  and  direct 
them   upon  an  8-inch  plano-convex 

condenser.  Used  for  back  lighting, 
special  effects,  and  particularly  on 
sets  where  the  general  lighting  must 
be  in  low  key. 

(22)  MR  Type  16  Cinelite.—A  spun 
aluminum  reflector,  finished  inside  by 
wire    brushing    and   chemical   treat- 
ment,   which    gives    it    a    diffusing 
characteristic.      Used     where    light 
portable  equipment  is  required. 

(23)  MR  Type  45  Rifle  Lamp.— 
Stamped  metal  reflector,  chromium- 
plated   with   rifled   corrugations   for 
diffusion        Used    for   general    floor 
lighting. 

(24)  MR  Type  220  18-Inch   Sun 

Spot. — An  18-inch  diameter  parabolic  glass  mirror  or  faceted  metal 
reflector,  with  spill  ring  as  standard  adjunct.  Used  for  general 
illumination,  for  back  lighting  and  cross  lighting  in  small  and 
moderate  size  sets,  and  for  projecting  light  into  back  areas  of 
sets. 

(25)  B&M  Type  T5  Studio  Spotlamp.— A  short-focus  Fresnel-type 
lens  in  front  of  the  bulb  and  a  small  fixed  spherical  mirror  behind  the 
bulb  project  light  forward  into  the  field.    This,  in  combination  with 
the  light  projected  around  the  lens  from  the  24-inch  reflector,  gives 
an  even,  intense  light.     For  the  large  mirror,  either  a  24-inch  diame- 
ter aplanatic  reflector  or  a  10-inch  focus  glass  mirror  is  used.     The 
aplanatic  reflector  produces  a  very  even  field  of  light.    Greater  pene- 
trating power  for  long  throws  may  be  obtained  with  the  parabolic 


Lamp    No.    26.    MR    type    226 
24-inch  studio  sun  spot. 


52 


STUDIO  LIGHTING  COMMITTEE  REPORT     [j.  s.  M.  p.  E. 


Lamp  No. 

28.  MR  type  206  baby  solarspot. 

29.  B  &  M  type  6  baby  keg-lite. 

30.  B  &  M  type  K  keg-lite. 


Lamp  No. 

31.  MR  type  208  solarspot. 

32.  MR  type  210  junior  solarspot. 

33.  MR  type  214  senior  solarspot. 


Jan.,  1939]  STUDIO  LIGHTING  COMMITTEE  REPORT  53 

glass  reflector.  Used  for  back  lighting,  cross  lighting,  front  lighting, 
and  effect  lighting. 

(26)  MR  Type  226  24-Inch  Studio  Sun  Spot.— A  24-inch  diameter 
glass  mirror,  with  a  spill  ring  that  allows  only  projected  light  to  leave 
the  lamp.    Used  for  back  lighting  large  sets,  in  which  case  the  heads 
are  removed  from  the  pedestals  and  are  mounted  on  parallels  or  plat- 
forms built  at  the  top  of  the  set  or  hung  from  the  stage  roof  or  ceiling. 

TABLE  in 

Incandescent  Lamps  for  Set  Lighting 

*Degrees  Beam  Bulb  Bulb 

Lamo  Divergencies  No.  No. 

No  Unit  Min.         Max.  **(B  &  W)      (Color) 

20  MR  36  Studio  Spotlamp  8  44  8 

21  MR  26  Studio  Spotlamp  8  44  8 

22  MR  16  Cinelite1*  60  60  16 

23  MR  45  Rifle  Lamp  60  60  4-5  15 

24  MR  220 18"  Sun  Spot  8  18  3-7  14 

25  B  &  M  T-5  Studio  Spotlamp"  8  40  2-3  13-14 

26  MR  226  24"  Sun  Spot  12  24  2  13 

27  B  &  M  24"  Sun  Spot  12  24  2  13 

28  MR  206  Baby  Solarspot11  8  40  9 

29  B  &  M  Baby  Keg-Lite  Type  6  6  45  9 

30  B  &  M  Keg-Lite  Type  K  4  44  3-7  14 

31  MR  208  Solarspot"  10  44  8 

32  MR  210  Junior  Solarspot11  10  44  3-7  14 

33  MR  214  Senior  Solarspot  10  44  2  13 

34  Sky  Light™  180  180  2  13 

35  Broadside  (Doubles)  90  90  5  15 

36  36"  Sun  Spot  12  24  1  12 

37  Overhead  Strip  Lamp  5  15 

*  Approximate  figures  referring  to  usable  photographic  light. 
**  For  black-and-white  photography. 

The  lamps  are  used  where  a  large  quantity  of  light  is  to  be  supplied 
by  a  small  number  of  units ;  for  front  lighting  on  deep  sets ;  for  cross 
lighting  where  high  contrast  is  desired  or  on  wide  sets  where  the 
camera  angle  requires  that  the  cross  light  be  projected  from  a  dis- 
tance ;  for  effect  lighting  such  as  in  simulating  sunshine  through  win- 
dows or  doorways,  or  interior  light  into  exterior  darkness  in  night 
shots ;  and  for  similar  special  requirements  demanding  beams  of  high 
intensity. 

(27)  B&M  24-Inch  Sun  Spot.— Similar  to  No.  26  in  design  and  use. 


54  STUDIO  LIGHTING  COMMITTEE  REPORT      [J.  S.  M.  P.  E. 

(28)  MR  Type  206  Baby  Solarspot.—A  6-inch  diameter  Fresnel- 
type  lens.    The  small  size  of  this  lamp  permits  its  use  in  places  where 
the  larger  lamps  can  not  be  accommodated,  particularly  where  it  is 
necessary  to  conceal  a  source  of  high-intensity  light. 

(29)  B  &  M  Baby  Keg-Lite  Type  6.— A  short-focus  6-inch  diameter 
Fresnel-type  lens  combined  with  a  pref ocused  mirror ;  used  for  special 
effects  and  where  small  units  are  required. 

(30)  B  &  M  Keg-Lite  Type  K— A  10-inch  diameter  by  6-inch  focus 
Fresnel-type  lens.     A  set  spherical  mirror  projects  the  rear  light  from 
the  bulb  toward  the  lens.     A  general  purpose  unit  within  its  intensity 

limits;  used  for  front  lighting,  back 
lighting,  and  modeling. 

(31)  MR  Type  208  Solar  spot. —An 
8-inch    diameter    Fresnel-type    lens. 
A  rhodium-plated  spherical  mirror  is 
used  at  the  rear  of  the  bulb  to  direct 
the  light  toward  the  lens.     Used  for 
back  lighting,  modeling,  and  general 
front    lighting    within    its    intensity 
range. 

(32)  MR  Type  210  Junior  Solar- 
spot. — A    10-inch   diameter   Fresnel- 
type  lens.    A  rhodium-plated  spheri- 

Lamp  No.  34.     Sky  light.  J  f 

cal  mirror  is  used  at  the  rear  of  the 

bulb  to  direct  the  light  toward  the  lens.    Used  for  back  lighting, 
front  lighting,  cross  lighting,  and  modeling  within  its  intensity  range. 

(33)  MR  Type  214  Senior  Solar  spot. — A  14-inch  diameter  Fresnel- 
type  lens.    A  rhodium-plated  spherical  mirror  is  used  at  the  rear  of 
the  bulb  to  direct  the  light  toward  the  lens.    Used  where  high- wat- 
tage units  are  desirable,  for  back  lighting,  front  lighting,  and  side 
lighting. 

(34)  Sky  Light. — A  shallow  diffuse  reflector  about  24  inches  in 
diameter.    Used  below  and  above  sky  backings  and  screens,  where  a 
flat  even  light  distribution  is  required. 

(35)  Broadside  (Doubles). — Two  flood:type  reflectors  housed  in  one 
unit,  used  for  floor,  side,  and  overhead  lighting:    One  of  the  first 
incandescent  units  made. 

(36)  36-Inch  Sun  Spot. — A  36-inch  diameter  glass  mirror.     Used 
where  the  highest  intensity  of  projected  light  is  required  from  an 
incandescent  tungsten  source. 


Jan.,  1939] 


STUDIO  LIGHTING  COMMITTEE  REPORT 


55 


(37)  Overhead  Strip  Lamp. — A  trough-like  unit  containing  sockets 
for  five  1000-watt  PS  52  bulbs.  Used  to  supply  fill-in  light  where 
it  is  difficult  to  use  a  more  bulky  housing. 


TABLE  IV 


Incandescent  Bulbs 


Nominal 

Bulb  Rated 

No.  Watts 


Nominal 
Amps. 


Bulb**  Volts  Amps.  Base 

"MP"  Type  Lamps  (for  Black-and-White  Photography) 


1 

10,000 

G-9613 

2 

5,000 

G-6413 

3 

*2,000 

G-4813 

4 

1,500 

PS-521* 

5 

1,000 

PS-5213 

6 

****!,  000 

T-2013 

7 

•1,000 

G-48 

8 

1,000 

G-40 

9 

****500 

T-2013 

10 

****500 

G-30 

11 

****200 

T-10 

110-115-120 
110-115-120 
110-115-120 
110-115-120 
110-115-120 
110-115-120 
110-115-120 
110-115-120 
110-115-120 
110-115-120 
110-115-120 

87.0 
43.5 
17.4 
13.1 
8.7 
8.7 
8.7 
8.7 
4.4 
4.4 
1.7 

Mog.  Bip. 
Mog.  Bip. 
Mog.  Bip. 
Mog.  Scr. 
Mog.  Scr. 
Mog.  Scr. 
Mog.  Bip. 
Med.  Bip. 
Medium*** 
Mog.  Scr. 
Medium 

"CP"  Type  Lamps  (for  Color  Photography — with  Proper  Filter) 

(All  "CP"  Type  Lamps  Operate  ar  3380°K.  Color  Temperature) 

12  10,000     G-9611  110-115-120    87.0     Mog.  Bip. 

13  5,000     G-64n  110-115-120    43.5    Mog.  Bip. 

14  2,000     G-4811  110-115-120     17.4    Mog.  Bip. 

15  2,000     PS-5211  105-120  17.4    Mog.  Scr, 

Additional  Types  Frequently  Used  in  Studio  Work 

16  1,000     PS-35™  105-120  8.7    Mog.  Scr. 

(No.  4  Photoflood) 

17  500     A-25™  105-120  4.4    Med.  Scr. 

(No.  2  Photoflood) 

18  250     A-21 13  105-120  2.2    Med.  Scr. 

(No.  1  Photoflood) 

*  Available  also  in  Mogul  screw  base  for  older  equipments. 
**  G  =  spherical,  PS  =  pear  shaped,  T  =  tubular,  A  =  modified  pear  shaped. 
Numbers  refer  to  diameter  in  1/8  inch. 

***  Some  units  require  the  med.  bip.  base,  others  the  med.  scr.  base  or  med. 
pref.  base. 

****  Used  in  utility  lamps,  lighting  fixtures,  table  and  floor  lamps. 


56 


STUDIO  LIGHTING  COMMITTEE  REPORT     [j.  s.  M.  p.  E. 

TERMS  USED  IN  STUDIO  LIGHTING  PRACTICE 


The  terms  applied  to  the  various  units  of  motion  picture  studio  lighting  equip- 
ment are  legion  and  vary  from  studio  to  studio,  and  even  from  month  to  month. 
Sometimes  a  lamp  is  described  by  its  type  number  alone;  or  by  the  rated  current 


Lamp  No. 

35.  Broadside  (doubles). 

36.  36-inch  sun  spot. 

37.  Overhead  strip  lamp. 

in  the  case  of  arc -spotlights;  or  by  the  kilowatt  rating  of  incandescent  units. 
In  some  instances  the  mirror  diameter  supplies  the  name.  Below  are  some  com- 
monly used  terms,  the  "Lamp  Numbers"  referring  to  the  preceding  sections: 


Term 
Broad 
Side  Arc 
Sixty-Five 
Ninety 
One-Seventy 
Twenty-Four 
Thirty-Six 
Eighty 

1000- Watt  Spot 
Rifle 
Eighteen 
T-5 


Lamp 
No. 

2-3-35 

2-3 

4 

5 

6 

7 

8 

9 

20-21-31 
23 
24 
25 


Lamp 

Term 

•        No. 

Twenty-Four  Inkie 

26-27 

5KW 

26-27 

Baby 

28-29 

Keg 

30 

Junior 

32 

Senior 

33 

Pan  or  Skypan 

34 

Doubles 

35 

10KW 

36 

Strip 

37 

Jan.,  1939]  STUDIO  LIGHTING  COMMITTEE  REPORT  57 

The  following  are  a  few  terms  used  for  material  and  equipment  associated  with 
the  use  of  studio  lamps: 

Silks. — Frames  equipped  with  china  silk  diffusers,  hung  on  the  fronts  of  lamps 
to  diffuse  the  light  and  reduce  the  intensity. 

Jellies. — Frames  equipped  with  chemically  treated  gelatin.  Used  for  the  same 
purposes  as  silks. 

Scrim. — Black  gauze  used  in  various  places  to  reduce  intensity  and  diffuse 
light. 

Diverging  Doors. — Strips  of  cylindrical  glass  lenses.  Used  on  Sun  Arcs  for 
light  diffusion. 

Snouts. — Various  shapes  of  black  sheet  metal  hangars.  Used  on  the  front  of 
lamps  to  block  out  undesired  light. 

Spill  Rings. — A  series  of  sheet  metal  tubes,  used  in  front  of  incandescent 
bulbs  in  mirror  type  lamps  to  block  off  angular  rays  emanating  from  the  front 
surface  of  the  bulb  filament  (see  photographs  of  lamps  24-26). 

Spot  Projector. — A  unit  equipped  with  a  condenser  system  that  fits  on  the  front 
of  a  Type  170  carbon  arc  lamp  in  place  of  the  Fresnel-type  lens;  used  to  produce  a 
sharply  defined  round  spot  of  light. 

Gobos,  Flags,  Cheese  Cutters,  Niggers,  Etc. — It  is  often  desirable  to  place  opaque 
screens  at  various  points  on  a  set  to  keep  all  or  a  part  of  the  light  from  reaching 
certain  areas  or  objects.  These  screens  are  painted  dull  black  and  are  rectangular, 
square,  or  circular,  as  the  occasion  may  require. 

LAMP  FILTERS  FOR  COLOR  PHOTOGRAPHY 

Carbon  Arc  Lamps. — Carbon  arc  lamps  1-2—3  are  used  for  Technicolor  pho- 
tography without  color  filters.  All  types  of  high-intensity  rotating  arc  lamps  re- 
quire a  Type  Y-l  straw  gelatin  filter.4 

Incandescent  Bulb  Lamps. — Where  incandescent  bulbs  are  used  on  Technicolor 
photography  a  special  blue  glass  filter  is  required  along  with  a  series  of  CP  Type 
bulbs,  which  burn  at  a  uniform  color  temperature  of  3380  °K.n 

REFERENCES 

(All  references  are  to  J.  Soc.  Mot.  Pict.  Eng.) 

1  MOLE,  P.:  "New  Developments  in  Carbon  Arc  Lighting,"  XXII  (Jan.,  1934), 
No.  1,  p.  51. 

2  HANDLEY,  C.  W.:   "Lighting  for  Technicolor  Motion  Pictures,"  XXV  (Nov., 
1935),  No.  5,  p.  423. 

3  RICHARDSON,  E.  C. :    "Recent  Developments  in  High-Intensity  Arc  Spot- 
lamps  for  Motion  Picture  Production,"  XXVIII  (Feb.,  1937),  No.  2,  p.  207. 

4  HANDLEY,  C.  W.:    "The  Advanced  Technicof  Technicolor  Lighting,"  XXIX 
(Aug.,  1937),  No.  2,  p.  169. 

5  JOY,  D.  B  ,  AND  DOWNES,  A.  C.:    "Characteristics  of  High-Intensity  Arcs," 
XIV  (March,  1930),  No.  3,  p.  291. 

6  JOY,  D.  B.,  BOWDITCH,  F.  T.,  AND  DOWNES,  A.  C.:    "A  New  White-Flame 
Carbon  Arc  for  Photographic  Light,"  XXII  (Jan.,  1934),  No.  1,  p.  58. 

7  BOWDITCH,  F.  T.,  AND  DOWNES,  A.  C.:   "The  Photographic  Effectiveness  of 
Carbon  Arc  Studio  Light-Sources,"  XXV  (Nov.,  1935),  No.  5,  p.  375. 


58  STUDIO  LIGHTING  COMMITTEE  REPORT     [j.  S.  M.  P.  E. 

8  BOWDITCH,  F.  T.,  AND  DOWNES,  A.  C. :    "The  Radiant  Energy  Delivered  on 
Motion  Picture  Sets  from  Carbon  Arc  Studio  Light-Sources,"  XXV  (Nov.,  1935), 
No.  5,  p.  383. 

9  BOWDITCH,  F.  T.,  AND  DOWNES,  A.  C.:    "Spectral  Distributions  and  Color- 
Temperatures  of  the  Radiant  Energy  from  Carbon  Arcs  Used  in  the  Motion 
Picture  Industry,"  XXX  (April,  1938),  No.  4,  pp.  400-409. 

10  RICHARDSON,  E.  C.:    "A  Wide-Range  Studio  Spotlamp  for  Use  with  2000- 
Watt  Filament  Globes,"  XXVI  (Jan.,  1936),  No.  1,  pp.  95-102. 

11  Report  of  the  Studio  Lighting  Committee,  XXX  (March,  1938),  No.  3,  p.  294. 

12  Report  of  the  Studio  Lighting  Committee,  XXV  (Nov.,  1935),  No.  5,  p.  432. 

13  FARNHAM,  R.  E.,  AND  WORSTELL,  R.  E. :    "Color  Quality  of  Light  of  Incan- 
descent Lamps,"  XXVII  (Sept.,  1936),  No.  3,  p.  260. 

C.  W.  HANDLEY,  Chairman 

R.  E.  FARNHAM  V.  E.  MILLER  G.  F.  RACKETT 

W.  C.  KUNZMANN  M.  W.  PALMER  E.  C.  RICHARDSON 

J.  H.  KURLANDER  F.  WALLER 

DISCUSSION 

MR.  GOLDSMITH:  Is  not  this  report  the  first  assembly  of  such  material  in  com- 
plete form? 

MR.  GEIB  :  Yes.  This  is  the  first  time  anyone  has  attempted  to  give  a  com- 
plete catalogue  of  studio  lighting  equipment. 

MR.  WOLF :     Is  the  mercury- vapor  arc  lamp  in  commercial  use? 

MR.  GEIB  :     No. 

MR.  WOLF:  I  understand  they  are  used  in  Holland,  in  combination  with 
sodium- vapor  lamps  to  get  a  more  balanced  spectrum. 

MR.  GOLDSMITH:  It  would  be  interesting  to  know  whether  that  type  of  com- 
bination could  be  used  for  color  photography  because  while  it  might  give  a  sub- 
jective effect  of  white  with  the  addition  of  sodium  vapor-lamps,  it  certainly  would 
not  give  the  physically  continuous  spectrum  of  an  arc  or  an  incandescent  lamp. 

It  would  therefore  be  interesting  to  know  whether  the  Technicolor  engineers 
could  use  a  combination  of  that  sort. 

MR.  WOLF  :  I  understand  that  the  combination  is  used  a  great  deal  in  Holland 
for  television  work  and  for  studio  work. 

MR.  GOLDSMITH  :  As  the  pressure  is  increased  in  the  mercury  lamps  the  back- 
ground spectrum  becomes  more  and  more  intense  and  a  certain  quasi-continuity 
of  the  spectrum  is  obtained. 

MR.  CARLSON:  That  is  correct.  As  the  operating  pressure  in  the  mercury  arc 
type  of  lamp  is  increased  the  continuity  of  the  spectrum  is  definitely  improved, 
together  with  an  increased  output  of  red  energy. 

In  the  case  of  medium-pressure  air-coole4  lamps  the  spectrum  is  still  largely  of 
the  discontinuous  or  band  type.  The  high-pressure  water-cooled  capillary  lamp 
now  on  the  market  shows  a  continuous  spectrum  superimposed  on  the  band 
spectrum.  For  still  higher  pressures  the  band  characteristics  largely  disappear. 
Thus  the  mercury  arcs  that  are  now  available  are  well  adapted  to  monochromatic 
photography,  but  not  for  color  work — nor  is  their  light  easily  filtered  because  of 
the  "humps"  in  the  energy  vs.  wavelength  curve.  Possibly  the  sensitivity  charac- 
teristics of  the  color  film  could  be  adapted  to  the  light.  Information  on  the  lamp 


Jan.,  1939]  STUDIO  LIGHTING  COMMITTEE  REPORT  59 

was  published  in  the  September,  1938,  issue  of  the  JOURNAL  by  Farnham  and  Noel. 

MR.  WOLF:  What  are  the  efficiencies  of  the  light-sources  now  as  compared 
with  what  they  were,  say,  several  years  ago? 

MR.  DOWNES  :  The  efficiency  of  light-sources  in  the  studios  is,  so  far  as  I  have 
been  able  to  learn,  not  of  great  importance.  What  is  wanted  is  a  steady  light- 
source,  and  one  that  can  be  directed  to  the  particular  part  of  the  set  with  cer- 
tainty and  assurance  that  it  is  going  to  continue  to  deliver  uniform  amount  and 
quality  of  light  during  the  time  the  photographing  is  done. 

The  efficiencies  in  lumens  per  watt  on  the  sets  in  the  studios  must  vary  through 
tremendously  wide  limits  because  of  the  fact  that  a  very  large  number  of  the  light- 
sources  used  are  focused  to  deliver  spots  of  various  sizes  on  the  set  and  as  a  result 
the  luminous  efficiencies  are  extremely  variable.  These  various  spot  units  can  be 
focused  to  deliver  spots  from  about  three  feet  in  diameter  to  very  large  ones,  and 
it  would  therefore  be  very  difficult  indeed  to  obtain  any  figures  for  lumens  per 
watt  except  with  a  bare  light-source  which,  considering  how  they  are  used,  would 
mean  little  or  nothing. 


PHOTOGRAPHIC  EFFECTS  IN  THE  FEATURE  PRODUCTION 

"TOPPER"* 


R.  SEAWRIGHT  AND'  W.  V.  DRAPER** 

Summary. — An  account  is  given  of  the  various  types  of  photography  used  in  the 
feature  production  "Topper."  Among  the  shots  discussed  are  a  split  screen  against 
a  projected  background,  demonstrating  the  feasibility  of  such  treatment.  Other  ef- 
fects are:  multiple  exposures,  animated  split  screen,  animated  travelling  mattes, 
straight  animation,  intricate  matching  of  action,  and  a  new  process  of  subtractive 
matting. 

A  statement  is  included  on  the  precautions  taken  to  eliminate  weave  between  the 
production  shots  taken  with  Mitchell  cameras  and  the  duping,  which  was  done  on 
Bell  &  Howell  machines.  The  paper  is  illustrated  with  various  selections  from  the 
picture,  made  by  the  processes  described. 

The  reel  witnessed  at  the  Washington  Convention  of  the  Society 
on  April  26,  1938,  contained  shots  from  the  Hal  Roach  production 
Topper  that  are  representative  of  the  various  types  of  trick  photog- 
raphy used  in  the  picture.  They  consist  mainly  of  multiple  expo- 
sures, animated  split  screen,  animated  travelling  mattes,  straight  ani- 
mation, intricate  matching  of  action,  and  subtractive  matting. 

Most  of  the  shots,  particularly  where  the  characters  appear  or  dis- 
appear, are  dupes  made  in  contact  on  an  optical  printer  with  hard 
mattes  in  the  optical  head.  The  general  procedure  on  the  set  was  to 
take  as  much  of  the  empty  set  as  was  needed,  either  before  the  scene 
was  started  or  after  it  was  finished,  depending,  of  course,  -upon  which 
end  of  the  scene  the  split  screen  was  to  be  used. 

The  camera  was  allowed  to  dolly  or  pan  either  before  or  after  the 
part  of  the  action  requiring  the  split  screen .  At  no  time  was  the  camera 
anchored.  Of  course,  extreme  care  was  taken  not  to  move  the 
camera  while  the  portions  of  the  scenes  were  being  shot  in  which  a 
character  was  to  appear  or  disappear. 

In  most  instances  the  action  was  taken  on  the  set  exactly  as  if 

*  Presented  at  the  1938   Spring   Meeting  at  Washington,  D.  C. ;    received 
Nov.  11,  1938. 

**  Hal  Roach  Studios,  Culver  City,  Calif. 

60 


PHOTOGRAPHIC  EFFECTS  IN  "TOPPER"  61 

there  were  no  split  screen  involved.  The  invisible  actor  either  oc- 
cupied the  position  in  which  he  would  ultimately  appear,  or  if  the 
visible  actor's  action  carried  him  too  close  to  or  past  the  spot,  the  in- 
visible one  would  run  in  and  occupy  his  position  as  soon  as  the  visible 
actor  was  sufficiently  clear  of  his  position.  The  invisible  actor  would 
then  be  cut  out  on  a  dupe  and  blank  set  substituted  in  his  position 
until  time  for  him  to  appear. 

The  last  three  scenes  in  the  reel  shown  at  Washington — of  the  two 
materializing  in  the  car  seat,  Miss  Bennett  materializing  on  the  bed, 
and  the  background  shot  driving  down  Broadway — for  various  rea- 
sons were  discarded,  but  were  inserted  in  this  reel  to  illustrate  further 
what  can  be  done  by  the  simple  treatment  previously  described. 

It  will  be  seen  from  the  preceding  that  there  was  nothing  particu- 
larly new  in  Topper.  It  was  made,  we  might  say,  by  doing  what  had 
to  be  done  by  the  best  available  system  known  to  the  operators  in 
charge.  It  is  almost  safe  to  say  that,  with  what  is  now  known  about 
what  can  be  done  in  the  handling  of  film,  there  is  a  way  of  solving  any 
problem  if  the  result  justifies  the  effort.  The  text  describing  the 
various  scenes  accompanies  the  appropriate  illustrations  on  the 
following  pages. 


62 


R.  SEA  WRIGHT  AND  W.  V.  DRAPER         [J.  S.  M.  P.  E. 


The  first  scene  on  the  reel 
(Fig.  1),  in  which  Constance 
Bennett  and  Gary  Grant  ma- 
terialize on  a  log,  is  a  combina- 
tion split  screen  and  lap  dissolve. 
After  Roland  Young's  action 
carried  him  to  the  left  half  of 
the  set,  the  screen  was  split 
optically  and  a  straight  shot  of 
the  background  substituted  in 
the  right  half  until  such  time  as 
it  took  for  Miss  Bennett  and 
Mr.  Grant  to  enter  and  take 
their  places.  The  matte  was 
then  dissolved  out  and  the 
original  scene  dissolved  in. 


FIG.  1. 


Jan.,  1939] 


PHOTOGRAPHIC  EFFECTS  IN  "TOPPER' 


63 


The  scene  showing  Mr.  Grant 
disappearing  as  he  approaches 
the  automobile  (Fig.  2)  is  a  lap 
dissolve  from  the  scene  showing 
him  walking  away  into  the  set. 
The  door  of  the  car  was  then 
opened  by  an  operator  inside  the 
car.  The  changing  of  the  tire 
(the  scene  following  that  shown 
in  Fig.  2)  involved  various  types 
of  animation,  manipulating  wires 
and  concealed  operators.  For 
instance,  the  jack  was  manipu- 
lated by  an  operator  in  a  pit 
under  the  car  and  the  car  let 
down  by  another  operator. 


FIG.  2. 


64 


R.  SEAWRIGHT  AND  W.  V.  DRAPER         [j.  s.  M.  P.  E. 


The  scene  in  which  Constance 
Bennett  "zips"  herself  out  (Fig. 
3)  was  projected  and  mattes 
animated  to  follow  the  action  of 
her  hand.  Miss  Bennett  got 
up  and  ran  off  the  set  as  soon 
as  her  action  was  finished  and 
Mr.  Young  held  his  position 
until  she  was  clear.  The  length 
of  film  necessary  to  get  her  off 
the  set  was  then  cut  out,  and  by 
use  of  the  animated  matte,  plain 
background  was  made  to  re- 
place Miss  Bennett  as  the 
"zipping"  action  progressed. 

The  shot  at  the  elevator 
where  Miss  Bennett  and  Mr. 
Grant  disappear  while  holding 
up  Mr.  Young,  was  a  simple 
lap  dissolve.  After  they  had 
decided  that  they  should  dis- 
appear, they  held  their  position 
for  a  sufficient  footage  to  cover 
the  dissolve,  then  releasing 
Young,  they  ran  off  the  set 
while  Young  continued  with  his 
action  of  swaying  back  and 
forth.  The  scene  was  then  dis- 
solved as  they  faded  away, 
shortening  the  amount  of  foot- 
age it  took  them  to  run  out  of 
the  scene.  Young's  action 
matched  up  and  he  was  dissolved 
back  in  again. 


FIG.  3. 


Jan.,  1939] 


PHOTOGRAPHIC  EFFECTS  IN  "TOPPER" 


65 


The  shot  of  Miss  Bennett  with 
the  vase  of  flowers  (Fig.  4)  was 
a  split  screen,  lap  dissolve,  and 
wire  shot.  Young  played  the 
scene  alone  until  after  the  box 
on  the  desk  had  been  moved 
with  wires,  after  which  Miss 
Bennett  entered  the  scene  and, 
taking  her  position  on  the  corner 
of  the  desk,  lifted  the  vase.  An 
operator  watched  the  action 
through  an  anchored  still  cam- 
era, following  the  action  of  the 
vase  and  marking  it  on  the 
ground-glass.  The  set  was  then 
cleared  and  from  the  same  set-up 
the  vase  was  lifted  with  wires  as 
closely  as  possible  in  the  path 
and  at  the  same  speed  as  Miss 
Bennett  had  lifted  the  vase. 
What  discrepancy  there  was 
between  the  two  actions  of  the 
vase  was  corrected  optically  on 
the  lavender  positive  print  and 
a  split  screen  dupe  made  imme- 
diately in  front  of  Young  in 
which  the  clear  set  with  the 
vase  on  wires  was  substituted 
until  the  vase  started  up  at 
which  time  the  set  was  dissolved 
out  and  Miss  Bennett  dissolved 


FIG.  4. 


66 


R.  SEA  WRIGHT  AND  W.  V.  DRAPER 


The  bit  of  feminine  apparel 
walking  by  itself  without  visible 
means  of  support  (Fig.  5)  was 
photographed  against  black  vel- 
vet on  a  girl  wearing  a  black 
velvet  suit.  The  shooting  con- 
tinued up  to  the  point  where 
they  were  snatched  off  and 
put  into  the  background  by  a 
rather  involved  process  known 
as  subtractive  matting,  in  which 
the  whole  of  a  developed  and 
fixed  print  is  converted  back  to 
silver  bromide  and  re-sensitized, 
after  which  the  background 
printed  into  the  heavy  deposit 
of  silver  representing  the  black 
velvet.  Snatching  the  pants  off 
was  a  case  of  matched  action. 
In  one  take — that  is,  in  the  one  in 
which  the  pants  walked — Roland 
Young  snatched  at  them  in  an 
empty  set.  From  the  same 
position  he  snatched  a  real  pair 
from  a  wire — and  the  scenes 
were  cut  in  action. 


FIG.  5. 


I 

_^~-  I  ^  g 


Perhaps  the  most  daring  shot 
from  a  standpoint  of  braving 
possible  technical  troubles  was 
the  shot  of  Miss  Bennett  mate- 
rializing in  the  roadster  (Fig.  6). 
This  was  a  split  screen  shot 
against  a  projection  background. 
Actually  no  difficulties  were  ex- 
perienced as  precautions  were 
taken  to  prevent  them.  The 
Mitchell  camera  taking  the  shot 
was,  of  course,  equipped  with 
precision  pins.  The  lavender 
positives  were  printed  tails  first, 
on  a  Bell  &  Howell  printer  which 
uses  the  same  perforations  for 
registry  as  the  Mitchell.  A 
special  shuttle  was  built  for  the 
optical  printer  registering  pins 
at  the  bottom,  so  that  through- 
out the  whole  process,  the  same 
perforations  were  used  for  regis- 
trv. 


FIG.  6. 


The  pen  writing  by  itself 
(Fig.  7)  in  the  close  shot  was 
straight  animation.  The  pen 
was  equipped  with  a  pin  in  the 
point  which  was  stuck  into  the 
blotter  holding  the  pen  upright 
as  the  animation  progressed. 
The  long  shot  was  done  with 
wires. 


FIG.  7. 


FIG.  8. 


The  shower-bath  sequence 
(Fig.  8)  was  a  composite  of  four 
shots.  Efforts  to  photograph  a 
person  in  a  black  suit  under  a 
shower  proved  surprisingly  un- 
convincing. The  effect  finally 
was  achieved  by  playing  several 
jets  of  air  against  the  water  in 
front  of  a  black  velvet  drop. 
The  steam  and  the  action  of  the 
soap  were  also  taken  against 
black  velvet  and  the  three  shots 
doubled  in  over  the  shower-bath 
set.  The  fixtures  were  worked 
from  the  opposite  side  of  the 
partition.  Miss  Bennett's  ma- 
terialization after  the  shower 
was  a  simple  split  screen  and 
lap  dissolve. 


The  following  shot,  however, 
in  which  she  snuffs  her  cigarette 
out  after  she  has  dematerialized 
(Fig.  9)  came  dangerously  near 
to  being  complicated.  She  was 
dematerialized  in  a  split  screen- 
lap  dissolve,  substituting  the 
empty  set  in  her  half  until  she 
had  time  to  run  out  of  the 
original  scene.  The  position  of 
her  cigarette  which  had  been 
located  on  a  still  camera  ground- 
glass,  was  matched  and  the 
cigarette  carried  down  to  the 
tray  with  wires.  This  bit  of 
action  then  had  to  be  substituted 
for  the  empty  set.  The  last 
puff  of  smoke  was  then  shot 
coming  through  a  hole  in  black 
velvet  and  doubled  in  where 
Miss  Bennett's  face  was  last 
seen. 


FIG.  9. 


PHOTOGRAPHIC  EFFECTS  IN  "TOPPER' 


69 


The  scene  in  which  Grant 
materializes  back  of  Eugene 
Pallette's  arm  (Fig.  10)  de- 
pended more  upon  acting  for  its 
success  than  upon  trick  photog- 
raphy. Pallette  struck  his  posi- 
tion in  the  alcove  and  held  it 
without  moving  while  Grant 
ran  in  and  took  his  place  behind 
Pallette's  arm.  After  allowing 
sufficient  footage  for  the  transi- 
tion, they  both  picked  up  the 
action  and  continued  the  scene. 
In  the  dupe  it  was  only  neces- 
sary to  shorten  the  scene  with  a 
lap  dissolve  the  amount  of  foot- 
age it  took  Grant  to  get  to  his 
place. 


FIG.  10. 


70 


R.  SEA  WRIGHT  AND  W.  V.  DRAPER 


[J.  S.  M.  P.  E. 


The  scene  in  the  cafe  where 
both  Miss  Bennett  and  Grant 
disappear  (Fig.  11)  is  what  tech- 
nically is  known  as  a  "head- 
ache," the  necessity  of  keeping 
the  background  action  consistent 
being  the  principal  problem.  In 
this  scene,  which  was  a  double 
split  screen-lap  dissolve,  instead 
of  dissolving  the  characters  into 
an  empty  background,  they  had 
to  be  replaced  by  people,  many 
of  them  moving.  Perhaps  it 
should  be  said  that  the  shot  was 
made  by  the  "perseverance" 
method. 


FIG.  11. 


Jan.,  1939] 


PHOTOGRAPHIC  EFFECTS  IN  "TOPPER" 


71 


The  stairway  scene  (Fig.  12) 
was  made  by  the  same  method 
with  slight  variations.  It  may 
have  been  noticed  that  Young 
crossed  to  Grant's  side  of  the 
screen  the  moment  Grant  dis- 
appeared. From  the  techni- 
cian's standpoint  it  was  a  mo- 
ment too  soon,  for  Mr.  Grant 
had  not  yet  run  off  the  set. 
As  a  consequence,  it  was  neces- 
sary to  animate  the  split  screen 
matte  which  had  been  intro- 
duced to  dissolve  Grant  out  of 
the  scene  ahead  of  Young  as  he 
advanced,  and  re-animate  the 
opposite  matte  which  printed  in 
the  plain  background  without 
losing  that  almost  imaginary 
line  that  makes  a  perfectly 
blended  match. 


FIG.  12. 


72 


R.  SB  A  WRIGHT  AND  W.  V.  DRAPER 


Grant's  sitting  on  the  chande- 
lier (Fig.  13)  was  a  split  screen- 
lap  dissolve.  The  only  diffi- 
culty was  that  the  closeness  of 
the  actors  below  necessitated  a 
very  sharp  blend  and  an  un- 
usually shaped  matte. 


FIG.  13. 


LATENT  IMAGE  THEORY  AND  ITS  EXPERIMENTAL 
APPLICATION  TO  MOTION  PICTURE  SOUND- 
FILM   EMULSION* 


W.  J.  ALBERSHEIM** 


Summary. — In  a  previous  paper  the  writer  attempted  to  show  that  the  latent  photo- 
graphic image  is  formed  in  two  distinct  and  separate  steps.  In  the  present  paper  this, 
theory  is  compared  with  recent  physical  research.  The  writer  concludes  that  each  of 
the  photographic  steps  consists  of  the  attraction  of  one  electropositive  silver  ion  to  a 
sensitizing  speck  on  the  grain  surface  which  has  previously  captured  an  electron. 

The  reciprocity  law  failure  at  high  intensities  is  explained  by  the  minimum  time  re- 
quired for  the  attraction  of  a  silver  ion.  The  reciprocity  law  failure  at  low  intensities 
is  explained  by  assuming  that  a  sensitizing  speck  which  has  attracted  only  one  silver 
ion  is  unstable  and  that  the  number  of  grains  activated  by  a  single  capture  decreases 
exponentially  with  time. 

From  these  physical  theories  the  writer  deduces  mathematical  relations  governing 
the  photographic  characteristics.  H&D  curves  and  reciprocity  failure  curves  com- 
puted from  these  relations  are  in  good  qualitative  agreement  with  experimental 
curves. 

The  assumption  of  an  unstable  intermediate  state  of  insufficiently  exposed  grains 
implies  certain  effects  of  delayed  fogging  and  delayed  development  which  are  verified 
by  experiments. 

In  a  previous  paper1  by  the  writer,  it  was  deduced  from  the  shape 
of  the  photographic  H&D  curves  that  a  grain  of  motion  picture  film 
must  be  hit  by  at  least  two  photons  in  order  to  become  developable. 
At  that  time,  no  physical  explanation  of  this  "double  hit"  theory  was 
attempted  and  hope  was  expressed  that  the  theory  would  be  supplied 
by  research  physicists. 

There  is,  of  course,  no  lack  of  research  nor  of  physical  theories  in 
this  field.  On  the  contary,  one  is  overwhelmed  by  an  ever-increasing 
wealth  of  literature ;  in  fact,  the  writer's  attention  was  called  to  some 
important  contributions  after  completion  of  the  analytical  studies 
which  form  the  basis  of  the  present  report. 

Unfortunately,  the  various  investigators  do  not  agree  even  in  some 
of  the  most  fundamental  assumptions.  While  most  of  them  assume 

*  Presented  at  the  1938  Spring  Meeting  at  Washington,  D.  C. 
**  Electrical  Research  Products,  Inc.,  New  York,  N.  Y. 

73 


74  W.  J.  ALBERSHEIM  [j.  s.  M.  P.  E. 

that  the  latent  image  consists  of  silver  atoms,  recent  work  carried 
out  in  the  Eastman-Kodak  Research  Laboratories2  suggests  that 
the  solarized  image  alone  forms  a  metallic  silver  cluster  which  is 
amenable  to  chemical-physical  development  only,  whereas  the  latent 
image,  being  subject  to  chemical  development,  must  be  differently 
constituted.  A  further  divergence  of  opinion  exists  with  regard  to 
the  number  of  photons  required  for  latent  image  formation.  Some 
of  the  earlier  research3  showed  that  in  single  halide  crystals  the  ratio 
of  metal  atoms  reduced  by  light  to  the  number  of  absorbed  photons, 
i.  e.,  the  photochemical  quantum -efficiency  lies  between  one-quarter 
and  one.  Other  workers  stated4  that  a  film  grain  in  a  photographic 
emulsion  must  absorb  about  100  quanta  in  order  to  become  develop- 
able, whereas  a  recent  series  of  tests5  proved  that  for  various  negative 
and  positive  emulsions  the  number  of  quanta  incident  upon  an  average 
sized  photographic  grain  needed  for  formation  of  a  latent  image  is 
of  the  order  of  50.  Since  the  grains  are  transparent  and  absorb  only 
a  fraction  of  the  incident  photons,  their  photographic  quantum-effi- 
ciency must  be  considerably  greater  than  l/w. 

We  begin  our  survey  of  experimental  knowledge  with  the  photo- 
graphic toe  characteristic.  As  shown  in  the  previous  paper,1  the 
shape  of  the  characteristic  curve  requires  for  the  activation  of  the 
film  grain  by  light,  two  separate  physical  steps  which  were  mathe- 
matically represented  in  that  paper  by  differential  equations  49  and 
72.  It  must  be  understood,  however,  that  these  two  steps  are  not 
necessarily  to  be  identified  with  two  single  photons  of  light.  They 
only  mean  two  processes  initiated  by  illumination  in  which  the  oc- 
currence of  the  second  process  depends  upon  the  completion  of  the 
first,  so  that  the  probability  (or  average  quantum-efficiency)  for  the 
formation  of  a  latent  image  by  a  photon  is  the  product  of  the  separate 
probabilities  for  the  two  steps  taken  singly.  The  absolute  number  of 
photons  required  for  the  completion  of  each  step  affects  the  photo- 
graphic inertia  rather  than  the  toe  shape  of  the  characteristic  which 
depends  only  on  the  ratio  between  the  2  efficiencies.  This  toe  shape 
at  low  exposures  is  difficult  to  analyze  from  ordinary  logarithmic 
H&D  curves.  It  has  been  investigated  by  many  authors  but,  in 
most  cases,  with  sources  of  illumination  quite  different  from  visible 
light. 

Silberstein  and  Trivelli6  as  well  as  Jauncey  and  Richardson7  found 
that  the  density  of  the  developed  photographic  image  originating 
from  weak  x-ray  exposures  grows  in  linear  proportion  to  exposure. 


Jan.,  1939]  LATENT  IMAGE  THEORY  75 

Silberstein  and  Trivelli  also  proved  that  the  number  of  developable 
photographic  grains  is  equal  to  the  number  of  photons  impinging 
upon  the  grain  surface.  For  these  x-rays  therefore  a  "single  step" 
theory  is  established.  However,  x-ray  photons  have  an  energy  con- 
tent which  is  many  thousand  times  greater  than  that  of  the  visible 
light  photons  used  in  sound  recording.  Since  a  double  step  process 
is  claimed  for  visible  light,  there  must  be  an  intermediate  range  of 
wavelengths  at  which  a  transition  from  the  single  to  the  double  step 
occurs.  This  seems  to  be  actually  the  case : 


FIG.  1.     Characteristics  of  film  exposed  to  x-rays  and  between 
double  intensifying  screens. 

Hirsh8  showed  that,  although  in  images  formed  by  hard  x-rays  the 
density  increases  proportionally  to  irradiation,  the  density-exposure 
characteristic  begins  to  curve  upward  when  the  x-rays  are  softened 
to  a  wavelength  of  over  six  Angstrom  units. 

This  curvature  means  that  the  probability  of  latent  image  forma- 
tion is  proportional  to  a  power  greater  than  one  of  irradiation  so  that 
on  the  average  more  than  one  photon  per  grain  is  needed  for  the  latent 
image  formation. 

In  order  to  free  the  comparison  between  x-rays  and  light  from  the 
influence  of  the  types  of  emulsion  used,  the  writer  asked  the  Eastman 
Kodak  Laboratories  to  supply  information  regarding  the  different 
photographic  characteristics  of  one  and  the  same  emulsion  when 


76 


W.  J.  ALBERSHEIM 


[J.  S.  M.  P.  E. 


exposed  to  x-rays  and  to  visible  light.  Mr.  Wilsey  of  the  Eastman 
Kodak  Physics  Department  very  kindly  supplied  the  characteristics 
which  are  shown  in  Fig.  1  of  this  paper.  Curve  A  of  this  figure  is 
produced  by  x-ray  exposure  and  shows  a  long  sloping  toe  which  in  the 
previous  paper1  was  shown  to  correspond  to  the  "single  hit"  theory 
and  incidentally  to  high  transparencies  of  the  emulsion.  Curve  B 
is  obtained  by  exposing  the  same  film  between  double  intensifying 
screens.  These  intensifying  screens  are  fluorescent  surfaces  which 
emit  a  great  number  of  visible  light  photons  when  hit  by  a  single 
x-ray  photon.  As  stated  by  Mr.  Wilsey  "when  the  exposure  is  made 
with  intensifying  screens,  practically  the  whole  photographic  effect  is 
due  to  the  fluorescent  light  from  the  screens  so  that  the  H&D  response 


PAR  SPEED    PORTRAIT    FILM 

DENSITY -i 


1.2          18         QA 

L06     INTENSITY 


FIG.  2.     Constant-density  curves  for  different  development  times. 
A — 5  minutes'  development,  B — 30  minutes'  development. 


is  essentially  that  due  to  exposure  to  light  on  both  sides  of  the  film." 
It  is  evident  that  the  toe  of  Curve  B  shows  a  much  greater  sharpness 
than  that  of  Curve  A  corresponding  to  an  exponent  greater  than  one 
and  therefore  consistent  with  the  double  hit  theory.  (A  three-step 
process  can  not  play  any  important  part  in  sound-film  emulsions  be- 
cause it  would  cause  a  toe-curvature  greater  than  that  found  under 
actual  conditions.) 

If  one  accepts  the  two  separate  steps  of  exposure  as  a  fact,  what  are 
the  known  properties  and  time  requirements  assignable  to  these 
steps  ?  This  information  may  be  derived  from  a  comprehensive  series 
of  tests  conducted  chiefly  by  Jones,  Webb,  and  other  physicists 
of  the  Eastman  Kodak  Laboratories  on  the  subjects  of  "Reciprocity 
Law  Failure"  and  "Intermittency  Effect,"  which  are  closely  linked 
to  each  other.  The  main  features  of  the  reciprocity  failure  effect 


Jan.,  1939] 


LATENT  IMAGE  THEORY 


77 


may  be  illustrated  by  our  Fig.  2  which  is  a  reproduction  of  Fig.  9 
of  a  paper  by  Jones  and  Hall  published  in  1929.9  In  this  figure 
the  logarithm  of  exposures  required  to  produce  given  densities  is 
plotted  against  the  logarithm  of  intensity.  The  reciprocity  law 
assumed  that  the  photographic  effect  depended  only  on  the  total 
number  of  photons  impinging  on  the  film  grain ;  that  is,  on  the  total 
exposure.  If  this  were  true  the  curves  of  Fig.  2  should  be  horizontal 
lines.  Actually,  the  lines  "fail"  to  be  horizontal  and  curve  upward 
at  very  low  and  very  high  intensities.  Kron  and  Halm10  re- 
ported that  this  curvature  can  be  approximated  by  a  catenary  rela- 
tionship, which  in  Fig.  2  is  illustrated  by  solid  lines.  However,  at 
the  left  side  of  this  figure  one  sees  dashed  lines  breaking  away  from 


Ea&imao  Slow  Lantern  P1a4es 


3.9        45        It        3.7        2.3        2.9         1.5        at         0.1         1.3         19        2.5        3.1 
LOG,  INTENSITY 

FIG.  3.  Constant-density  curves  for  slow  emulsion  showing  low- 
intensity  departure  from  catenary  equation.  Top  curve,  density  = 
4.2;  bottom  curve,  density  =  0.20. 

the  catenary  and  rising  at  an  angle  of  about  45°  which  limits  the  curve 
fitting  range  of  the  catenary  and  deprives  it  of  physical  significance. 
Another  series  of  such  curves  is  shown  in  Fig.  3  which  is  a  reprint  of 
Fig.  11  of  the  above-mentioned  paper.  In  these  figures,  the  logarithm 
of  intensity  is  used  as  abscissa  axis  in  accordance  with  historical 
precedent.  This  historical  usage  seems  to  the  writer  to  be  an  un- 
fortunate choice  which  has  for  a  long  time  beclouded  the  underlying 
physical  relations.  When  one  talks  of  a  greater  intensity  in  a  physical 
process,  such  as  a  baseball  hit,  one  thinks  of  greater  speed  or  greater 
muscle  tension.  But  when  one  talks  of  intensity  of  illumination  with 
light  of  a  given  color,  all  the  little  baseballs,  that  is,  the  photons  of 
light,  hit  their  objective  with  the  same  speed  and  with  the  same 
energy  of  impact.  What  is  meant  by  "intensity  of  illumination" 
is  actually  the  number  of  photons  per  second,  and  denotes  a  quantity 


78  W.  J.  ALBERSHEIM  [j.  s.  M.  P.  E. 

rather  than  a  quality.  The  physical  dimension  of  this  "intensity" 
is  energy  times  sec."1.  Since  the  scale  is  logarithmic,  it  is  only 
necessary  to  reverse  the  sides  of  the  diagram  in  order  to  plot  as 
abscissa  axis  the  logarithm  of  the  reciprocal  factor;  that  is,  of  the 
average  time  interval  between  successive  photon  impacts  upon  a  film 
grain. 

This  slight  difference  in  the  interpretation  of  the  abscissa  axis 
might  have  greatly  speeded  up  the  progress  of  research,  for  the  im- 
pact time  interval  has  recently  been  shown  by  J.  H.  Webb5  to  be  a 
most  important  factor  in  the  reciprocity  law  failure  effect.  In  his 
investigation  of  the  relation  between  reciprocity  failure  and  the  in- 
termittency  effect,  Webb  discovered  that  intermittent  exposures  are 
equivalent  to  continued  exposures  with  an  equal  total  number  of 
photons  radiated  and  equal  total  duration,  provided  that  the  interrup- 
tion cycle  is  completed  within  the  average  time  interval  between 
successive  photon  impacts  upon  one  and  the  same  grain  surface. 

Webb  defines  reciprocity  law  failure  as  the  effect  produced  by  the 
tune  distribution  of  quanta  reception  by  a  photographic  grain  (in 
agreement  with  our  double  step  theory) . 

This  theory  accounts  for  the  reduced  efficiency  at  both  extremes  of 
intensity,  or  rather  of  exposure  time  in  the  following  way :  Reciproc- 
ity failure  at  high  intensities  means  that  the  first  step  requires  for 
its  completion  a  small  but  definite  average  time  interval,  before  the 
second  step  can  take  place:  If  a  new  photon  hit,  or  group  of  hits, 
occurs  before  the  "step,"  i.e.,  the  physical  process  initiated  by  the 
previous  photon  hit  (or  hits)  has  had  time  to  be  completed,  the 
additional  hits  just  "do  not  count"  and  very  short  exposures  can  not 
utilize  all  received  light  quanta  for  the  production  of  developable 
photographic  grains. 

Disregarding  for  simplicity's  sake  the  statistical  variations  in  the 
time  requirements  of  the  first  step,  we  note  for  incorporation  into  our 
mathematical  picture,  that  the  effective  time  interval  between  successive 
photon  impacts  upon  a  film  grain  exceeds  the  actual  time  interval  by  a  fixed 
minimum  time  which  hereafter  is  called  the  "blocking  time." 

In  order  to  account  for  the  reduced  efficiency  of  the  photographic 
process  at  very  low  intensities,  that  is,  very  long  exposure  times,  it 
is  necessary  to  make  an  additional  assumption:  The  configuration 
produced  by  the  first  step  of  latent  image  formation  must  be  elec- 
trically or  chemically  unstable ,  a  stable  latent  image  being  only  ob- 
tained by  the  completion  of  the  second  step.  The  simplest  form  of 


Jan.,  1939]  LATENT  IMAGE  THEORY  79 

this  assumption  is  that  grains  activated  by  completion  of  the  first  step 
fade  back  to  the  unexposed  state  according  to  an  exponential  time  func- 
tion, as  if  the  activated  grains  were  a  radioactive  substance  re-emitting 
the  stored  light  energy  in  random  manner.  Some  of  this  released  energy 
may  be  detectable  by  photographic  or  other  methods.  The  time  after 
completion  of  exposure  in  which  the  number  of  activated  grains  is 
reduced  by  a  factor  e  will  be  introduced  into  our  equations  as  the 
"fading  time." 

However,  before  attempting  the  mathematical  analysis  of  this  de- 
layed step-by-step  mechanism,  an  attempt  should  be  made  to  find  a 
plausible  physical  explanation  for  the  somewhat  involved  process  of 
image  formation  which  we  deduced  from  two  such  well  established 
every-day  characteristics  as  the  H&D  curve  and  the  reciprocity  failure 
curve! 

As  previously  mentioned,  the  research  of  Przibram,  Smakula, 
Hilsch,  and  Pohl3  shows  that  in  single  silver  halide  crystals  the  photo- 
chemical action  consists  of  a  liberation  of  silver  atoms  from  their 
crystal  bonds  to  the  halide  ions,  the  number  of  atoms  thus  reduced 
being  proportional  to  the  number  of  photons  absorbed.  This  sug- 
gests that  the  two  photographic  steps  are  related  to  the  number  of 
deposited  silver  atoms  rather  than  to  the  number  of  incident  photons. 
This  interpretation  is  made  more  probable  by  the  above-mentioned 
fact  that  one  and  the  same  emulsion  has  entirely  different  character- 
istics when  exposed  to  x-rays  and  to  visible  light.  The  powerful  x-ray 
photons  blast  the  required  number  of  silver  atoms  out  of  the  halide 
crystal  in  a  single  hit,  whereas  the  weaker  light  photons  can  only  dis- 
place them  one  at  a  time.  The  double  step  hypothesis  is  thus  nar- 
rowed down  to  a  "double  atom"  hypothesis.  It  implies  two  claims 
which  must  be  substantiated :  (a)  That  in  spite  of  the  low  quantum 
efficiency  of  grain  exposure,  two  silver  atoms  deposited  at  the  right 
place  are  necessary  and  sufficient  to  make  a  photographic  grain  de- 
velopable; and  (b)  that  the  deposition  of  these  two  atoms  proceeds 
in  separate  steps. 

Considerable  light  is  shed  upon  the  minimum  size  of  development 
nuclei  by  the  research  of  W.  Reinders  and  his  associates.11  These 
investigators  condensed  extremely  thin  films  of  silver  on  glass  plates 
and  developed  them  in  solutions  containing  a  mixture  of  chemical 
developing  agents  and  free  silver  salts.  The  minimum  developable 
silver  density  turned  out  to  be  1/5ooth  of  that  corresponding  to  a  single 
atomic  layer.  The  authors  assumed  that  the  deposited  silver  atoms 


80  W.  J.  ALBERSHEIM  Q.  s.  M.  P.  E. 

combine  into  groups  if  they  are  condensed  within  a  mutual  distance, 
no  larger  than  that  vvhich  separates  them  in  a  metallic  silver  crystal, 
and  they  computed  the  probability  of  various  group  sizes.  The  ob- 
served minimum  density  was  found  just  sufficient  to  permit  the  oc- 
currence of  four-atom  groups.  Hence  it  was  concluded  that  aggre- 
gates of  four  or  more  silver  atoms  can  serve  as  nuclei  of  development. 

It  has  been  well  established  by  the  work  of  Sheppard  and  others12 
that  development  starts  at  so-called  sensitizing  specks  which  seem 
to  consist  of  silver  sulfide  molecules.  Reinders  found  that  sulfide 
molecules  can  act  as  centers  of  development  just  as  well  as  silver 
atoms.  Reinders  and  his  associates  then  applied  their  probability 
calculus  of  group  formation  to  the  amount  of  silver  sulfide  present 
in  photographic  emulsions.  They  found  that  according  to  Shep- 
pard's  reports  there  are  several  hundred  silver  sulfide  molecules  avail- 
able for  each  grain;  that  each  grain  is  likely  to  have  on  its  surface 
(and  perhaps  in  its  interior)  several  groups  consisting  of  two  sulfide 
molecules  each;  but  that  on  only  a  small  percentage  of  grains  (less 
than  10  per  cent)  aggregates  of  three  sulfide  molecules  will  be  formed. 
Now  one  may  "put  two  and  two  together:"  If  two  units  of  aggrega- 
tion are  supplied  by  the  sulfide  molecules  of  the  sensitizing  specks  and 
if  aggregates  of  four  are  needed  to  make  a  grain  developable,  it  is 
evident  that  two  additional  units  must  be  supplied  by  the  photo- 
graphic process  in  the  form  of  silver  atoms  or  silver  ions. 

The  manner  in  which  the  silver  atoms  find  their  way  to  the  sensitiz- 
ing specks  is  explained  by  J.  H.  Webb13  and  by  Gurney  and  Mott.14 
According  to  quantum  mechanics  the  primary  action  of  the  photon 
consists  in  knocking  loose  an  electron  from  its  attachment  to  a 
halide  ion.  If  the  electron  receives  sufficient  kinetic  energy  it  can 
travel  freely  through  the  previously  insulating  halide  crystal  as  if 
the  crystal  were  a  metal.  These  electrons  are  "captured"  at  regions 
of  the  crystals  in  which  the  electrical  potential  is  more  positive  than 
average.  Such  trapping  points  may  consist  of  small  fissures  and  ir- 
regularities in  the  crystal  or,  more  likely,  of  impurities,  such  as  the 
silver  sulfide  molecules  of  the  sensitizing  specks,  which  will  be 
charged  up  to  a  higher  negative  potential  by  the  captured  electron. 

Gurney  and  Mott  suggest  that  at  normal  temperatures  the  thermal 
agitation  will  throw  some  silver  atoms  out  of  their  normal  positions 
in  the  crystal  lattice,  producing  slightly  mobile  negative  "holes"  and 
and  highly  mobile  positive  silver  ions.  The  free  silver  ions  are  elec- 
trostatically attracted  by  the  electrons  captured  at  the  sensitizing 


Jan.,  1939]  LATENT  IMAGE  THEORY  81 

specks  and  move  toward  them  like  the  ions  in  a  liquid  electrolyte. 
This  description  fits  very  nicely  into  our  step-by-step  pattern.  The 
union,  or  at  least  close  association,  of  a  silver  ion  and  an  electron 
completes  one  photographic  step  by  neutralizing  the  free  charges. 
Since  the  net  motion  of  the  silver  ion  in  the  electrostatic  field  of  an 
electron  is  relatively  slow,  it  becomes  plausible  that  the  completion 
of  this  photographic  step  takes  a  measurable  amount  of  time.  Before 
neutralization  of  the  charges  the  negative  potential  of  the  captured 
electron  repels  any  further  electrons  which  might  be  liberated  by 
light  and  prevents  them  from  reaching  the  same  sensitizing  speck. 
This,  as  discussed  above,  accounts  for  the  reciprocity  failure  at  high 
intensities. 

After  completion  of  the  first  step  the  silver  ion  is  not  fully  stabilized 
at  the  sensitizing  speck.  Due  to  thermal  agitation  there  exists  during 
each  time  interval  a  small  probability  that  the  electron  which  at- 
tracted it  is  thrown  off  or  "evaporated."  In  that  case  the  electron 
begins  to  travel  freely  through  the  crystal  and  is  either  captured  by 
another  sensitizing  speck  or  by  a  crystal  irregularity,  or  it  may  re- 
combine  with  the  positive  electron  hole  created  when  it  was  first 
knocked  out  of  the  crystal  by  the  impact  of  the  photon.  This  leaves 
a  single  silver  ion  unattached  and  it  in  turn  resumes  a  random  thermal 
motion  through  the  crystal,  probably  being  recaptured  by  a  negative 
hole  in  the  lattice.  If,  however,  two  steps  are  completed,  that  is, 
two  silver  ions  and  their  electron  mates  are  united  at  the  sensitizing 
speck,  they  presumably  form  a  silver  molecule  which  is  more  stable 
than  a  single  silver  atom  and  much  less  likely  to  give  up  an  electron 
by  heat  evaporation.  By  assigning  silver  atoms  to  the  small  nuclei 
of  chemical  development  this  mechanism  'differs  from  the  view  of 
Evans  and  Hanson2  who  attribute  metallic  nature  only  to  the  solarized 
nucleus.  Possibly  the  discrepancy  can  be  solved  by  assuming 
that  in  small  nuclei  containing  very  few  silver  atoms,  these  atoms 
remain  closely  embedded  in  the  crystal  structure  so  that  chemical 
development  can  spread  out  from  the  sensitizing  speck  into  the 
crystal.  When  the  number  of  silver  atoms  grows  due  to  over- 
exposure,  they  exert  an  increasing  mechanical  pressure  upon  the  sur- 
rounding crystal.  Finally  this  pressure  will  tear  the  silver  cluster 
loose  from  the  silver  halide.  The  cluster  can  be  physically  developed 
by  silver  deposition  but  it  is  physically  and  chemically  divorced  from 
the  halide  crystal  which  reverts  to  the  "unexposed"  state  minus  a 
sensitizing  speck. 


82 


W.  J.  ALBERHSEIM 


[J.  S.  M.  P.  E. 


The  above  physical  interpretation  follows  closely  the  viewpoint  of 
Gurney  and  Mott  and  differs  from  it  mainly  by  assuming  that  the 
deposition  of  two  silver  atoms  suffices  to  make  a  grain  developable, 
whereas  the  said  authors  imply  that  a  greater  number  is  necessary. 
If  two  atoms  only  are  required  and  if  a  single  photon  has  sufficient 
energy  to  liberate  an  atom,  why  is  it  that  on  the  average  dozens  and 
in  some  emulsions  even  hundreds  of  photon  impacts  strike  a  grain 
surface  before  it  becomes  developable  ?  The  reason  may  be  found  in 
a  relatively  great  number  of  sensitizing  specks  per  film  grain.  Assume, 
for  instance,  that  the  grain  contains  ten  sensitizing  specks  of  which 


FIG.  4. 


Showing  the  reduction  of  photographic  efficiency  produced 
by  an  increase  in  the  intensity  of  light. 


nine  are  located  in  the  inaccessible  interior  of  the  grain  and  only  one 
on  the  surface  where  it  can  be  reached  by  the  developer.  Even  if 
there  were  no  other  trapping  possibilities  but  the  sensitizing  specks, 
the  probability  for  a  liberated  silver  ion  to  reach  a  sensitizing  speck 
on  the  surface  of  the  grain  would  be  only  Vioth.  The  probability  of 
a  second  silver  ion's  reaching  the  accessible  sensitizing  speck  pre- 
viously reached  by  the  first  ion  would  also  be  Vioth,  so  that  the  com- 
bined probability  for  the  formation  of  a  developable  grain  would  be 
Viooth  for  two  photon  impacts,  or  1/2ooth  per  photon. 

The  above-described  mechanism  accounts  for  the  known  experi- 
mental facts  in  a  qualitative  manner.  As  a  next  measure,  therefore, 
it  was  brought  into  a  simplified  mathematical  form.  The  differential 


Jan.,  1939] 


LATENT  IMAGE  THEORY 


83 


equations  governing  the  two  steps  and  their  solutions  are  given  in 
the  mathematical  appendix  attached  to  this  paper.  The  equations 
differ  from  the  double  hit  equations  of  the  previous  paper1  by  the 
blocking  time  required  between  the  first  and  second  step  and  by  the 
gradual  decay  of  film  grains  in  which  the  first  step  only  was  com- 
pleted. At  the  limit  of  zero  blocking  time  and  infinite  decay  time 
which  is  approached  for  medium  intensity,  the  new  equations  coincide 
with  the  old  ones.  The  H&D  curve  at  this  medium  intensity  has 
been  plotted  as  the  extreme  condition  in  the  curves  of  Figs.  4  and 
5.  Fig.  4  shows  the  reduction  of  photographic  efficiency  produced  by 


FIG.  5.     Computed  H&D  curves  for  long  exposure  times  at  decreas- 
ing intensities. 

an  increase  in  the  intensity  of  light.  At  extremely  high  intensity 
levels  shown  at  the  right  side  of  Fig  4,  the  shape  of  the  H&D  curves 
approaches  that  of  the  x-ray  characteristic,  Curve  A,  in  Fig  1.  One 
reason  for  this  is  that  the  efficiency  is  at  a  minimum  at  the  front  sur- 
face of  the  emulsion  where  the  excess  of  photons  is  greatest  so  that 
there  is  little  difference  between  the  photographic  effect  in  front  and 
rear  of  the  emulsion :  the  effective  penetration  increases.  Further- 
more, due  to  the  abundant  supply  of  freed  electrons  a  second  electron 
will  reach  every  sensitizing  speck  and  initiate  the  second  step  im- 
mediately after  completion  of  the  first  step.  This  makes  the  photo- 
graphic process  a  function  of  the  first  step  only,  i.  e.,  the  equivalent  of 
a  single-step  process.  The  time-scale  gamma  increases  with  intensity 


84 


W.  J.  ALBERSHEIM 


[J.  S.  M.  P.  E. 


due  to  the  high  penetrating  power,  and  the  straight-line  portion  of 
the  H&D  curve  is  shortened. 

In  Fig.  5,  a  similar  series  of  H&D  curves  has  been  computed  for 
long  exposure  times  at  decreasing  intensities.  The  photographic 
effect  falls  off  first  at  the  rear  of  the  emulsion  where  a  long  time  in- 
terval between  successive  photon  impacts  allows  the  end  products  of 
the  first  step  to  fade  away  before  a  second  photon  is  received.  Con- 
sequently, the  effective  penetrating  power  decreases,  producing  at 
extremely  low  intensities  a  reduced  gamma  and  a  straight-line  portion 
of  the  H&D  curve  extending  high  up  toward  the  shoulder.  Whether 
this  idealized  relation  is  followed  under  practical  conditions  may  be 


LOG  X  LOG   RELATIVE  MTCNSITY 

FIG.  6.     Constant  density  curves,  computed:    "reciprocity  failure." 

doubted.  The  great  effective  absorption  requires  extreme  exposures 
in  the  front  of  the  emulsion  to  obtain  a  reasonable  overall  density 
and  this  may  lead  to  solarization  and  loss  of  density  in  the  shoulder 
region. 

All  characteristics  of  Figs.  5  and  6  are  "time-scale"  curves.  The 
intensity-scale  curves  deviate  increasingly  from  these  for  the  more 
extreme  exposure  ranges:  As  shown  in  the  Appendix,  the  ratio  of 
time-scale  gamma  to  intensity-scale  gamma  is  a  function  of  the  slope 
of  the  reciprocity  failure  curve. 

By  picking  points  of  equal  density  on  the  various  curves  of  Figs.  5 
and  6  and  plotting  their  exposure  logarithm  as  function  of  the  in- 
tensity logarithm,  one  obtains  "reciprocity  failure  curves"  as  shown 
in  Fig.  6.  Comparing  these  curves  with  the  experimental  curves  of 
Figs.  2  and  3,  one  notes  the  similar  character  although  the  rise  of 


Jan.,  1939] 


LATENT  IMAGE  THEORY 


85 


the  computed  constant  density  curves  at  the  high  and  low  ends  is 
somewhat  too  sudden.  The  shape  of  the  measured  reciprocity  failure 
curves  can  be  explained  by  considering  that  actual  film  emulsions 
do  not  have  grains  of  uniform  size  and  composition  as  assumed  in 
our  simplified  calculations,  but  that  they  consist  of  a  wide  range  of 
grain  sizes  containing  different  numbers  of  sensitizing  specks.  In  a 
smaller  grain  it  will,  on  the  average,  take  less  time  for  the  silver  ion 
to  reach  the  electrified  sensitizing  speck,  hence  its  blocking  time  is 
shorter.  On  the  other  hand,  the  statistical  fluctuations  are  smaller 
in  a  smaller  grain,  thus  reducing  the  probability  for  electrons  to 
evaporate  from  the  sensitizing  speck  and  increasing  the  decay  time. 


*i* 

FIG.  7.     Constant  density  curves,  computed :    "reciprocity  failure. " 

Hence,  actual  reciprocity  failure  curves  will  be  produced  by  the  super- 
position of  a  great  number  of  contributing  curves  which  are  trans- 
posed laterally.  There  will  therefore  be  on  each  side  of  the  curves  an 
intermediate  range  of  reduced  and  fairly  constant  slope,  depending 
upon  the  statistical  distribution  of  grain  sizes. 

In  Fig.  6,  lines  of  equal  exposure  time  have  been  drawn,  in  the 
form  of  thin  straight  lines  rising  at  an  angle  of  45  degrees  toward  the 
right  in  a  manner  which  seems  to  have  first  been  used  by  Jones  and 
Webb.15  Since  the  reciprocity  failure  is  determined  by  two  time 
constants,  namely,  the  blocking  time  and  the  decay  time,  it  would 
be  more  instructive  to  plot  the  logarithm  of  the  total  irradiation  (It) 
as  a  function  of  log  T  rather  than  of  log  /.  This  has  been  done  on 


86 


W.  J.  ALBERSHEIM 


[J.  S.  M.  P.  E. 


Fig.  7.  It  is  seen  that  the  photographic  process  is  completely  ineffi- 
cient at  exposure  times  shorter  than  the  blocking  time.  At  somewhat 
longer  exposure  times  the  required  irradiation  reaches  a  minimum, 
and  at  extremely  long  exposure  times,  exceeding  the  decay  time,  the 
required  irradiation  for  constant  densities  increases  with  the  square 
root  of  exposure  time. 

Having  thus  verified  that  even  in  its  simplified  mathematical  form 
the  two-step  theory  fits  the  facts  that  led  to  its  adoption,  we  con- 
sidered it  necessary  to  put  it  to  an  experimental  test  by  checking  new 


A:  AT"  contrast  without  fogging. 

B:  2AE. 

C:  AE-1000  cycle  single  amplitude. 

D:  JEj-D.C.  component  of  signal. 

G:  E2-Foggmg. 

H:  AT"  contrast  with  fogging. 

/:  Ei  +  £2. 

FIG.  8.     Increase  of  contrast  by  fogging. 

facts  which  can  be  predicted  from  it.  The  most  significant  of  these 
facts  seems  to  be  connected  with  the  claimed  instability  of  the  activa- 
tion produced  by  the  first  photographic  step.  In  the  under-exposed 
toe  region  of  the  characteristic  the  irradiation  produces  relatively 
very  few  grains  in  which  the  two  steps  of  latent  image  formation  are 
completed,  but  a  much  greater  number,  in  which  one  step  only  has 
taken  place,  that  is,  one  silver  atom  transported  to  a  sensitizing  speck 
on  the  grain  surface.  After  development  these  under-exposed  pic- 
tures show  negligible  contrast.  However,  their  contrast  can  be  in- 
creased by  superimposing  to  the  picture  a  constant  illumination. 


Jan.,  1939]  LATENT  IMAGE  THEORY  87 

This  is  known  as  "fogging,"  and  depending  on  whether  the  super- 
imposed d-c.  exposure  takes  place  before  or  after  the  under-exposed 
modulated  exposure,  one  speaks  of  "pre-fogging"  and  "post-fogging." 
The  purpose  and  effect  of  fogging  are  illustrated  by  Fig.  8.  According 
to  our  view,  fogging  can  only  be  efficient  if  the  time  interval  between 
the  two  superimposed  exposures  is  smaller  than  the  decay  time  so 
that  the  activated  "single-step"  grains  do  not  have  time  to  fade  out. 
This  theory  was  verified  by  the  following  "fading  test":  A  sound- 
track was  exposed  to  a  thousand-cycle  signal  with  very  low  intensity 
of  illumination,  E\  (Fig.  8)  producing  a  specular  density  of  only 
0.07.  Upon  this  weak  signal  exposure  we  superimposed  a  uniform 
fogging  exposure  Ez  with  about  three  times  greater  average  light  in- 
tensity. One  part  of  this  fogging  exposure  was  applied  ten  minutes 
before  the  signal,  a  second  part  one-half  hour  after  the  signal,  and 

TABLE  I 

Fading  Test 

Hours  between  Relative  Level  Specular 

Signal  and  Fogging  (Db.)  Density 

No  fogging  -17.4  0.069 

-0.2  0.0  0.168 

+0.5  0.0  0.172 

+  1.0  -  0.05  0.164 

2.0  -  0.2  0.165 

4.0  -  0.5  0.160 

6.0  -  0.75  0.158 

21.5  -  2.1  0.142 

further  parts  after  increasing  time  intervals  up  to  21.5  hours.  The 
result  of  this  test  is  tabulated  in  Table  I  and  shown  on  Fig.  9.  It 
shows  that  fogging  applied  within  ±0.5  hour  of  the  signal  improves 
the  a-c.  output  by  17  db.  and  that  with  increased  time  intervals  this 
benefit  was  reduced  by  about  0.1  db.  per  hour.  During  the  test  time 
of  21  hours,  the  total  average  density  decreased  from  0.17  to  0.14. 
This  decrease  in  density  would  have  increased  the  level  by  0.6  db.  if 
the  modulation  had  remained  constant.  The  actual  decrease  of 
modulation  therefore  amounts  to  about  2.7  db.  Since  without  fog- 
ging the  signal  level  was  negligible,  the  modulation  loss  can  only  be 
explained  by  a  fading  of  the  partly  exposed  grains  in  the  time  in- 
terval between  signal  and  fog  exposures. 

The  change  of  average  density,  however,  may  have  a  double  ex- 
planation :    In  addition  to  the  fading  effect  proved  by  the  signal  loss, 


W.  J.  ALBERSHEIM 


[J.  S.  M.  P.  E. 


there  may  exist  an  increase  of  the  quantity  of  fully  exposed  grains 
available  for  development  with  time.  Such  an  "intensification  effect" 
has  been  mentioned  in  the  literature.  In  the  discussion  of  a  paper  on 
dry  hypersensitization  before  this  Society,16  Mr.  J.  I.  Crabtree  of  the 
Eastman  Kodak  Company  stated,  "It  is  well  known  that  you  may  get 
effective  hypersensitization  or  growth  of  the  latent  image  by  merely 
storing  the  latent  image."  The  two-step  theory  leads  one  to  expect 
such  an  intensification  from  the  following  reasoning:  Assume  that 
a  film  grain  has  ten  sensitizing  specks  and  that  due  to  an  insufficient 


-1.6  - 

-Z 


#et*rtve  tevei. 

8  •'  artcui**  Otff»trr~ 


0        Z         4         6         6        tO        fZ        14       16        19       20       ZZ 
HOURS  BeTHKCN  5IGWU.  GHD  fOSOIHS  CXPOSUKE 

FIG.  9.     Fading  test;     E.  K.  emulsion  1359. 

exposure  five  of  these  specks  have  received  one  silver  atom  each,  but 
that  in  none  of  these  specks  the  latent  image  has  been  completed  by 
the  absorption  of  a  second  silver  atom.  According  to  Gurney  and 
Mott's  theory,  some  of  these  specks  will  give  up  electrons  by  evapora- 
tion and  these  free  electrons  will  move  through  the  crystal  just  as 
if  they  had  been  liberated  by  an  additional  photon  of  light.  There 
exists,  therefore,  a  certain  probability  that  one  of  these  electrons  will 
be  captured  by  an  "activated"  sensitivity  speck  which  has  already 
received  one  silver  atom.  The  capture  of  this  additional  electron 
and  the  subsequent  attraction  and  absorption  of  a  second  silver  ion 
will  complete  the  second  step  for  this  grain  and  make  it  stable  and 


Jan.,  1939] 


LATENT  IMAGE  THEORY 


developable.  This  process  is,  of  course,  much  more  likely  to  occur 
after  weak  exposures,  when  the  number  of  partly  exposed  grains 
exceeds  that  of  fully  exposed  grains.  The  intensification  process  will 
occur  mainly  in  the  toe  region  of  the  H&D  curve.  In  a  modulated 
sound-track  exposed  in  the  straight-line  region  of  the  H&D  curve, 
the  effect  will  be  a  considerable  darkening  of  light  portions  and  a 
small  darkening  of  the  darker  portions,  that  is,  a  net  loss  of  modula- 
tion. 

This  theory  was  put  to  the  test  in  the  following  manner :    Alternate 
sections  of  film  were  exposed  with  unmodulated  light,  500-cycle 


CURVE    A   :  MODVl.»TION  LOSS 
6  ••  DCHSITY  INT 


4         5         6          7 
DAYS  BEFOB£  D£.Y£L,OPMeNT 


FIG.  10.     Intensification  test;     E.  K.  emulsion  1359. 

modulation  and  with  8000-cycle  modulation,  and  portions  of  all  three 
types  of  exposures  were  developed  after  storing  times  increasing  from 
1  to  13  days.  The  comparison  of  high-frequency  and  low-frequency 
signals  was  included  because  it  had  been  suspected  that  some  of 
the  electrons  liberated  in  the  fading  process  might  have  sufficient 
energy  to  expose  an  adjacent  grain  and  thus  to  produce  an  image 
diffusion  recognizable  as  a  high-frequency  loss.  The  results  of  this 
test  are  given  in  Table  II  and  Fig.  10.  After  correcting  for  the  daily 
variations  in  development  characteristics,  there  was  an  increase  in  the 
average  density  of  the  modulated  as  well  as  of  the  unmodulated  sound- 
tracks amounting  to  about  0.01  per  day  and  a  loss  of  modulation 
amounting  to  about  0.35  db.  per  day.  The  small  increase  in  the  den- 


90  W.  J.  ALBERSHEIM  [j.  s.  M.  P.  E 

sity  of  the  unmodulated  sound-tracks  shows  that  the  loss  of  modula- 
tion must  be  nearly  entirely  due  to  an  intensification  of  the  low-den- 
sity portions.  These  findings  explain  why  the  motion  picture  studio 
operators  dislike  recording  films  on  the  last  day  of  the  working  week 
and  storing  them  over  the  week-end  before  development. 


TABLE  II 

Intensification  Test 

Days  between                         Relative  Modulation  Average 

Exposure                                              (Dh>.)  Modu-  Average 

and  De-  lation  Specular 

velopment                          500~                         8000~  Loss  (Db.)  Density 

1                                       0.0                        -    0.7  0.0  0.62 

5                          -1.4                 -  8.3  1.35  0.69 

8                           -2.4                 -  9.4  2.4  0.71 

13                          -5.1                 -11.8  4.95  0.725 


Fortunately,  the  fading  time  of  commercial  emulsions  at  normal 
temperatures  is  so  long  that  the  photographic  result  is  not  noticeably 
impaired,  provided  development  takes  place  within  a  day  of  exposure. 
For  professional  motion  picture  work,  it  does  therefore  not  seem  neces- 
sary to  include  the  time  loss  of  modulation  due  to  low-density  in- 
tensification into  the  analytical  expressions  for  latent  image  formation 
which  are  cumbersome  enough  without  this  added  complication! 
From  the  experimental  point  of  view,  however,  the  positive  result  of 
the  fading  and  intensification  tests  has  encouraged  us  to  present  the 
"step-by-step"  or  "two-atom"  hypothesis  as  a  small  contribution 
to  the  comprehensive  quantum  theory  of  the  photographic  image 
formation  built  up  by  so  many  research  workers  and  culminating  at 
the  present  time  in  the  work  of  Webb  and  of  Gurney  and  Mott. 

REFERENCES 

1  ALBERSHEIM,  W.  J. :  "Mathematical  Relations  between  Grain,  Background 
Noise,  and  Characteristic  Curve  of  Sound-Film  Emulsions,"  /.  Soc.  Mot.  Pict. 
Eng.t  XI  (Oct.,  1937),  No.  4,  p.  434. 

2  EVANS,  R.  M.,  AND  HANSON,  W.  T.:  Phot.  J.  (Aug.,  1937),  p.  497. 

3  PRZIBRAM,  K.:   Z.  fur  Physik,  20  (1923),  p.  196. 
SMAKULA:  Z.fur  Physik,  59  (1929),  p.  603. 

HILSCH,  R.,  AND  POHL,  R.  W.  i  Z.fur  Physik,  64  (1930),  p.  607. 

4  EGGERT,  J.,  AND  NODDACK,  W.:  (1932).     See  ref.  14,  p.  156. 

6  WEBB,  J.  H.:   /.  Opt.  Soc.  Amer.,  23  (May,  1933),  No.  5,  p.  157. 
6  SILBERSTEIN,  L.,  AND  TRiVELLi,  A.  P.  H. :   Communication  No.  409,  Eastman 
Kodak  Research  Lab.;  Phil.  Mag.  (7th  Ser.),  9  (1930),  p.  787. 


Jan.,  1939]  LATENT  IMAGE  THEORY  91 

7  JAUNCEY,  G.  E.  M.,  AND  RICHARDSON,  H.  W.:    /.  Opt.  Soc.  Amer.  (May, 
1934),  p.  125. 

8  HIRSH,  F.  R. :  /.  Opt.  Soc.  Amer.  (Aug.,  1935),  p.  229. 

9  JONES,  L.  A.,  AND  HALL,  V.  C.:   Proceedings  7th  Internal.  Cong.  Phot.  (July, 
1928). 

10  KRON,  E.:  Pub.  Astrophys.  Obs.  Potsdam  (1913),  No.  67. 
HALM,  J.  R.:  Astron.  Soc.  Monthly  Notices  (June,  1922),  p.  473. 

11  REINDERS,  W.,  AND  HAMBURGER,  L.:    Z.  fur  Wissensch.  Phot.,  31  (1933), 
Nos.  1  and  2;  Ibid.,  No.  10. 

REINDERS,  W.,  AND  DE  VRIES,  R.  W.  P.:   Z.  fur  Wissensch.  Phot.,  56  (1937), 
Nos.  9  and  10,  p.  985. 

12  SHEPPARD,  S.  E.,  TRIVELLI,  A.  P.  H.,  AND  LOVELAND,  R.  P.:    J.  Franklin 
Inst.,  200  (1925),  p.  51. 

13  WEBB,  J.  H.:  J.  Opt.  Soc.  Amer.,  26  (Oct.,  1936),  No.  10,  p.  367. 

14  GURNEY,  R.  W.,  AND  Moxx,  N.  F. :   Proc.  Royal  Soc.,  164A  (Jan.,  1938),  p. 
151. 

15  JONES,  L.  A.,  AND  WEBB,  J.  H.:     "Reciprocity  Law  Failure  in  Photo- 
graphic Exposure,"  /.  Soc.  Mot.  Pict.  Eng.,  XXIII  (Sept.,  1934),  No.  3,  p.  142. 

16  DERSCH,  F.,  AND  DURR,  H.:    "A  New  Method  for  the  Dry  Hypersensitiza- 
tion  of  Photographic  Emulsions,"  /.  Soc.  Mot.  Pict.  Eng.,  XXVIII  (Feb.,  1937), 
No.  2,  p.  186. 


ANALYTICAL  APPENDIX 
COMPUTATION  OF  PHOTOGRAPHIC  CHARACTERISTICS 

(1)    Physical  assumptions  and  definitions 

(1.1)  A  latent  image  is  formed  by  the  deposition  of  two  or  more 
silver  ions  at  a  sensitizing  speck  located  on  the  grain  surface  and  ac- 
cessible to  the  developing  agent.    A  grain  in  which  this  image  forma- 
tion is  completed  is  called  "exposed." 

(1.2)  Silver  ions  are  transported  to  the  sensitizing  specks  by  the 
electrostatic  attraction  of  electrons  liberated  by  photons  from  the 
halide  crystal  and  trapped  by  the  sensitizing  speck;  this  transporta- 
tion takes  a  measurable  average  time  called  the  "blocking  time,"  tb. 
The  union  of  silver  ion  and  electron  neutralizes  the  free  electron 
charge  and  completes  a  photographic  "step." 

(1.5)  Before  completion  of  the  photographic  step  the  electron 
charge  repels  any  further  electrons  and  prevents  them  from  being 
trapped  by  the  same  sensitizing  speck. 

(1.4)  A  grain  in  which  only  one  photographic  step  is  completed, 
is  in  an  unstable  "activated"  state.  An  electron  and  subsequently 
the  silver  ion  attracted  by  it  may  be  lost  by  thermal  agitation.  The 
time  constant  of  this  loss,  i.  e.,  the  time  in  which  the  number  of  acti- 
vated grains  is  reduced  by  the  factor  e,  is  called  the  "fading  time,"  7}. 


92  W.  J.  ALBERSHEIM  [j.  s.  M.  P.  E. 

(1.5)  An  "exposed"  grain  per  1.1,  is  stable  and  remains  developable. 
(The  fact  that  an  excessive  amount  of  deposited  silver  atoms  may 
"solarize"  the  grain  and  make  it  inert  to  chemical  development,  is  of 
no  importance  in  the  exposure  range  of  motion  picture  sound-films.) 

(1.6)  The  grains  are  distributed  at  random  throughout  the  photo- 
graphic emulsion. 

(1.7)  Due  to  absorption  the  light  intensity  decreases  nearly  ex- 
ponentially with  depth  of  penetration.     The  absorption  constant  n 
is  defined  as  the  reciprocal  of  the  depth  at  which  the  number  of 
photons  per  second  is  decreased  by  the  factor  e. 

(1 .8)  Effective  Electron  Time  Interval. — The  average  time  interval  be- 
tween photon  impacts  on  the  grain  may  be  called  TP.     The  absorp- 
tion factor  of  single  grains  for  photons  is  called  pa.     The  probability 
that  an  electron  is  liberated  by  the  absorbed  photon,  is  called  pe. 
The  probability  that  a  liberated  electron  penetrates  to  an  unexposed, 
accessible  sensitizing  speck  is  called  pffl.      The  probability  that  a 
liberated  electron  penetrates  to  an  activated  accessible  sensitizing 
speck,  is  called  pgz. 

One  finds  the  effective  time  interval  between  photo-electrons  ar- 
riving at  an  unexposed  speck : 

(1} 

and  the  effective  time  interval  between  photo-electrons  arriving  at 
an  activated  speck 

CTi  W 


(1.9)  Additional  Symbols  (See  list  of  symbols  at  the  end  of  ap- 
pendix).— In  order  to  maintain  a  connection  with  the  previous  paper1 
we  define  g  as  the  fraction  of  the  grains  at  a  given  depth  in  the  emul- 
sion which  has  been  activated  but  not  fully  exposed,  and  r  as  the  fully 
exposed  fraction.  Where  convenient,  we  use  the  fading  factor: 

/  =  i  w 

and  the  intensity  factor: 


Jan.,  1939]  LATENT  IMAGE  THEORY  93 

(2)    Differential  equations  for  the  steps  of  latent  image  formation 

(2.1)  Equation  for  the  First  Step. — The  gross  increase  of  activated 
grains  is  proportional  to  the  available  number  of  unexposed  grains 
and  to  the  intensity  factor.    From  this  gross  increase  one  must  deduct 
the  decrease  due  to  fading  and  the  decrease  due  to  a  transformation 
of  activated  grains  into  fully  exposed  grains.    Hence: 

dt       Ti  Tf      dt 

or  g'  =  (1  -  r  -  f)  i  -  fg  -  r'  (6) 

(2.2)  Equation  for  the  Second  Step. — The  increase  of  exposed  grains 
is  proportional  to  the  available  number  of  activated  grains,  divided 
by  the  effective  electron  time  interval  plus  the  blocking  time : 

Hence  g  =  r'  (tz  +  tb)  (8) 

and  g'  =  r"  (t2  +  tb)  (9) 

(2.3)  Solution  for  a  Single  Emulsion  Layer. — Introducing  the  values 
of  8  and  9  into  6  one  finds : 

r  +  Ur'  +  Vr"  =  1  (10) 

with     U  =  Ti  +  (h  +  tb)(l  +  fTi)  (11) 

and     V  =  (tt  +  tb)Ti  (12) 

A  solution  of  10  must  have  the  form: 

r  =  1  +  ri-kit  +  rt-k*t  (13) 

At  /  =  0,  both  r  and  g  and,  in  view  of  (7),  r'  must  equal  zero.    Hence 
13  can  be  transformed  into: 

ri  +  r2  +  1  =  0  (14) 

riki  +  r2k2  =  0  (15) 

from  this  one  finds : 

r,    =  r^-r  (16) 

k\    —    K2 

r 2  -  -       ^  (17) 

and  r  =  1  +  _A_e-^        ***-** 


94  W.  J.  ALBERSHEIM  [j.  s.  M.  P.  E. 

From  10  one  finds  for  ki  and  kz  the  relation: 

rlf-kit  (i  _  Uki  +  W)  +  r2€-*rf(l  -  Ufa  +  VW}  =  0        (19) 
from  which  it  follows  that  : 

1  -  Uki  +  VkS  =  1  -  Ukt  +  VW  =  0  (20} 

I     (21} 


General  Equation  of  H&D  Curve.  —  Equations  18  and  21  de- 
scribe the  formation  of  the  latent  image  at  uniform  intensity,  and 
therefore  at  one  given  depth  in  the  emulsion.  The  itensity  itself  de- 
creases exponentially  with  depth  : 

iv  =  ioe~uy  (22} 

r<  =  I   =   Tioe»v  (23} 

Tz  =  C  Ti  =  5  =   T2oe"«  (24} 

If  R  is  defined  as  the  fraction  of  all  grains  in  the  emulsion  which  has 
become  developable,  one  sees  that  after  uniform  development  : 

R  =  D/Dm  (25} 

where  D  is  the  measured  density  and  D^,  the  highest  density  obtain- 
able with  the  same  emulsion  and  development.  According  to  defini- 
tion, R  is  found  as  : 


^  r 

Y  I 

Jo 


'y 
r  dy  (26} 


in  which  dy  denotes  a  differential  emulsion  layer  and   Y  the  total 
emulsion  thickness.    In  view  of  22: 

dy  =  -^  (27} 

*--4  ftr£- _4_ f*««  w 


in  which  r  is  the  transparency  of  the  emulsion  for  the  photographically 
active  light.  In  order  to  make  more  clear  which  factors  of  21  are 
functions  of  i,  it  may  be  transformed  into : 


Jan.,  1939]  LATENT  IMAGE  THEORY  95 

Combining  IS,  25,  and  28,  one  finds: 

D  =  -  .£       '    1-+       *L_  .-I-  -        *L-  .-*-     d4          (30} 


-  .£  f  '( 

lnrJiQT  \ 


Equations  29  and  30  constitute  the  general  solution  for  the  H&D 
curve  as  function  of  the  light  intensity  at  the  surface  and  the  length 
of  exposure,  assuming,  however,  that  all  grains  are  uniform  in  size 
and  constitution. 

(2.5)  Evaluation  of  H&D  Characteristics.  —  If  one  introduces  the 
full  values  of  k  from  29  into  30,  the  integral  becomes  very  compli- 
cated and  unmanageable  for  computation.  It  can,  however,  be  sim- 
plified by  the  following  considerations  : 

The  constant  C  in  equations  2  and  29  denotes  the  decrease  in  prob- 
ability of  electron  capture  by  the  presence  at  a  sensitizing  speck  of 
one  or  more  silver  ions  and  electrons  which  neutralize  each  other's 
charges.  The  additional  silver  atoms  may  lower  the  work  function 
and  thus  increase  the  probability  of  capture  so  that  C  would  be  some- 
what smaller  than  one.  However,  its  magnitude  will  not  differ  much 
from  unity  and  its  effect  will  be  about  equivalent  to  a  mere  change  of 
film  speed.  C  will  therefore  be  considered  equal  to  one  in  the  follow- 
ing computations. 

Furthermore,  blocking  time  and  fading  time  are  of  very  different 
orders  of  magnitude  :  The  blocking  time  is  measured  in  microseconds, 
the  fading  time  in  hours.  One  can  therefore  regard  the  one  as  zero 
or  the  other  as  infinite,  according  to  the  intensity  range  explored. 
This  splits  the  computation  into  a  high-intensity  and  a  low-intensity 
range. 

(2.51)  Computation  of  High-Intensity  Curves.  —  Equation  29  is  sim- 
plified into  : 


(31} 


Introducing  the  new  variable  : 

tbi  =  x  one  finds:  (32) 


ky  =  0.5* 


+  x      I  +  x 

=  i  (34} 

—L-  (55) 

1  +  x 


96  W.  J.  ALBERSHEIM  [j.  s.  M.  P.  E. 

*yi  +  *    _  i 


*  -  *yi  + 


(36) 


^ 


-  1  +  x  i          *  ( Q7\ 

i  _  in  +  *  -    -5-         ~  * 

Introducing  the  variable : 

t/tb  =  z  one  finds:  (38) 


X  \  X 

In  view  of  32 


-=  and 

x        ^ 


(42) 


xz  =  it  =  e  (43) 

(2.511)  Approximate  Solution  at  Relatively  Low  Intensities. — At  low 
intensities 

lim  x >  0  (44) 

The  integral  of  42  can  be  rewritten : 

x*€    L  *J 

For  small  x  values  this  approaches : 


For  a  given  exposure  time : 


™  =  ™  =  ^      hence  (47) 

x        i        e 

=  1  +  JL  Ce  e-e  (l+e)?Z  and 


-fct J> -•-'?. 

Equation  45  is  identical  with  equation  75  of  the  previous  paper  and 
confirms  that  at  low  intensities  the  blocking  time  does  not  have  any 
influence  upon  the  speed  or  the  shape  of  the  H&D  curve. 


*e-»*?  (50) 


Jan.,  1939]  LATENT  IMAGE  THEORY  97 

(2.512)  Approximate  Solution  at  Extremely  High  Intensities.  —  At 
high  intensities  lim  x  —  >  °°  .  This  reduces  42  to  : 

U-l  +  JL    f 

^Tjr 

D  =  Dm  (1  -  e-*)  =  Dm(l 

Equation  51  is  a  function  of  z  alone,  indicating  that  no  matter  how 
greatly  one  increases  the  intensity,  a  minimum  time  proportional  to 
the  blocking  time  is  required  for  the  formation  of  a  latent  image. 

The  reciprocity  failure  curve  is  determined  by  the  fact  that  t  be- 
comes independent  of  i.  Hence  : 

£MJ.  =  0  (52) 

a  log  i 

dlog  e  _  d  log  t  +  d  log  i  _  1  /^ 

d  log  i  d  log  i 

Equation  53  defines  the  reciprocity  failure  curve  as  a  straight  line 
rising  at  an  angle  of  45  degrees. 

The  shape  of  the  H&D  curve  for  extreme  intensity,  as  expressed 
by  51,  can  be  interpreted  by  comparing  equation  51  with  equation 
50  of  the  previous  paper:1  51  is  identical  with  the  function  resulting 
from  a  single  step  process  in  a  single  "layer"  of  emulsion  or  in  a  com- 
pletely transparent  emulsion.  That  is,  the  H&D  curve  at  extreme 
intensities  takes  the  shape  of  an  x-ray  characteristic. 

(2.51  3)  Strict  Solution  of  High-Intensity  Equation  .  —  Equation  42  can 
not  be  completely  solved  in  analytical  form.  By  partial  integration, 
however,  it  can  be  stripped  down  to  residual  terms  of  the  form  : 


This  function  was  discussed  and  used  in  the  previous  paper. *  (See 
equation  59,  p.  431,  and  Fig.  5  of  that  paper.) 

Even  with  this  abbreviating  symbol,  the  solution  for  R  or  D  re- 
mains rather  complicated : 

D   =  ~    —         -  '-    l+X          rx 

e 

— —  e       !+*    —  Z  (Fe   —   Fre)    —  (1    —  z) 

*--"-+**,   -  vi  + 

(55) 
From  this  equation,  the  curves  of  Fig.  4  were  computed. 


98  W.  J.  ALBERSHEIM  [j.  s.  M.  P.  E. 

(2.514)  Formulas  for  Gamma  at  High  Intensity. — The  function  for 
time-scale  gamma  is  nearly  as  cumbersome  as  equation  55  because 
equation  42  does  not  become  integrable  by  differentiating  with  regard 
to*: 

yt   =  (RDoo)  =  Dm  — =  0.434  Dm  — -—  (56) 

d  log  t  d  log  z  dz 

•\i* 

VT  <57> 

The  intensity-scale  gamma,  however,  can  be  freed  of  the  integral  sign : 


— • 


(2.52)  Computation  of  Low-Intensity  Curves. — As  discussed  in  the 
computation  of  high-intensity  curves,  the  constant  C  of  equations  2 
and  28  is  assumed  to  approximate  the  value  one.  Furthermore,  the 
blocking  time  is  regarded  as  infinitely  short  compared  to  the  long 
exposure  times.  Thus  equation  28  is  transformed  into  : 

kv  =  i  +  0.5/  ±  V(i  +  0.5/)2  -  i*  =  i  +  0.5/  ±  Vfi  +  0.25/2   (55) 
The  following  new  variables  are  introduced : 

i/f  =  x  (60) 

ft  =  z  (61) 

2  =  k/f  (62) 
This  transforms  18  into: 


r  =  1  + 

with  q  =  x  +  Vs  =»=  Vx  +  l/4  (64) 

(2.521)  Approximate  Solution  at  Relatively  High  Intensities. — lim 

63  can  be  written  in  the  form : 

r   =   1   +  e~  (3i+22)z/2  I   2? c-(gi-<72)z/2   _ 

L  2i  -  22 

^-l-e+(3l-«z2W2j  (65) 


Jan.,  1939]  LATENT  IMAGE  THEORY  99 

This  approaches  the  value : 

,  =  1  +  «-[V52-  *  •-  V*  -  ^t-1  e+VJ,]  (66) 

and,  due  to 

*Vx  <  1, 

r  =  1  -  €-«  (e  +  1) 

^=_    1 
^  In  r 

D  =  RDm  =^=  ~  [e-*r  -  €-e  -  Fe  +  Fer]  (70) 

This  equation  is  again  identical  with  equation  79  of  the  previous 
paper,  confirming  that  at  high  intensities  the  blocking  time  does  not 
influence  speed  or  shape  of  the  H&D  curves. 

(2.522)  Approximate  Solution  at  Extremely  Low  Intensities. — Under 
these  conditions, 

lim  x  ->•  0  (71) 

2l  =  1  +  *  =  1  (72) 

22  =  x*  (73) 

r  =  1  -  e~*2z  =^  1  —  €~'  with  (74) 

j  =  x*z  (75) 


I?    =  -!_    CX   r  <^   =  1          f     r 

\nrJXT       x    ~         2lnrjj^       j 


(76) 


The  density  becomes  a  function  of  j  alone.    The  reciprocity  failure 
curve  is  determined  by  the  equation : 

j  =  constant  (78) 

d  (long  x2)  +  d  log  z  =  0  (73) 

2d  (log  i)  =  d  (log  0  =  0  (80) 

d  (log  e)  =  d  (log  *)  +  d  (log  0  =  -  d  log  t  (&Z) 

The  reciprocity  failure  curve  becomes  a  straight  line  sloping  downward 
at  an  angle  of  45  degrees. 

The  shape  of  the  H&D  characteristic  77  approaches  that  of  a  single 
step  process,  but  with  a  low  transparency,  equal  to  the  square  of  the 
actual  transparency  r.  Accordingly,  one  sees  in  Fig.  5  that  the  toe  of 


100  W.  J.  ALBERSHEIM  [j.  s.  M.  P.  E. 

the  extreme  right  curve  becomes  rounded,  but  that  the  straight-line 
portion  extends  way  up  toward  the  shoulder. 

(2.523)  Strict  Solution  of  Low-Intensity  Equation.  —  We  have  : 

R  =  -  j_  r  **  (  i  +  _IL_  «-«.  —  «L_  €-522  )    (82} 

In  T  Jrx  x   \  qi  -  q2  ?i  ~  §2  / 


__i_,!.  (    } 

In  T  J  rx  x  qi  —  qz 

In  the  first  integral  of  83  substitute  q\  =  m*  (84) 
One  finds: 

x  ^  mz  —  m  (85) 

dx  •=  2m  -  1  (£0) 

32  =  (m  -  I)2  (87) 

In  the  second  integral  of  83  substitute  q.  =  »2  (88) 

One  finds  : 

x  =  n*  +  n  (89) 

dx  =  2n  +  1  (90) 

<Z2  =  (n  +  I)2  (W) 
This  transforms  83  into  : 


/0.5  +V0.25  +  a: 
^lic-«^ 
m 
5+V0.25  +  TX 


These  simplified  integrals  can  not  be  solved  completely.  But  by 
partial  integration  they  can  be  stripped  to  integrals  of  the  above- 
mentioned  type  F(e)  and  to  probability  integrals  defined  as : 


which  can  be  found  in  tables. 
The  solution  becomes : 


Jan.,  1939]  LATENT  IMAGE  THEORY  101 

in  which  equation  : 

ai  =  Vz  (0.5  4-  V0.25  +  *)  (95} 

a2  =  Vz(0.5  +  V0.25  +  r  *)  (96) 

b,  =  V7(-0.5  +  V0.25  +  *)  (57) 

&2  =  Vz  (-0.5  +  \/0.25  +  rx)  (98) 

(2.524)  Gamma  at  Low  Intensity.  —  The  time-scale  and  intensity- 
scale  gammas  can  be  found  by  differentiation  of  equation  82  with 
respect  to  log  /  and  log  i,  in  a  manner  analogous  to  that  shown  under 
2.514.  The  calculations  have  been  omitted  from  this  appendix  be- 
cause in  motion  picture  sound  recording  the  exposure  range  is  gen- 
erally not  in  the  low-intensity  part  of  the  reciprocity  failure  curve. 

(3)    Equations  of  Reciprocity  Failure  Curve 

The  concepts  of  blocking  time  and  fading  time  were  introduced 
into  the  theory  in  order  to  account  for  the  reciprocity  law  failure.  It 
will  therefore  be  shown  by  the  following  derivation  that,  and  how, 
the  reciprocity  failure  curve  is  determined  by  the  H&D  curve  29. 

For  a  given  emulsion  and  development,  Dm,  r,  tb,  and  Tf  are  con- 
stants. k  is  a  function  of  iv,  and  the  integral  29,  a  function  of  iQ  and 
/  only.  (The  suffix  of  i0  will  be  omitted  below  where  no  misunder- 
standing is  possible.) 

Reciprocity  curves  are  plotted  with  log  e,  that  is,  log  (it)  as  ordinate 
and  log  i  as  abscissa.  The  density  is  held  constant  for  any  given 
curve.  Since  log  i  and  log  /  are  the  only  independent  variables  the 
constancy  of  D  is  expressed  by  : 


d  log  e  =  d  log  (it)  =  d  log  i  +  d  log  /  (101) 


Equation  102  is  the  required  relation  between  ordinate  and  abscissa 
of  the  reciprocity  failure  curve. 


102                                     W.  J.  ALBERSHEIM  [j.  s.  M.  p.  E. 

(4)    Reciprocity  Curve  and  Intensity  —  to  Time-Scale  Gamma  Ratio 

In  the  "straight-line"  region  of  the  H&D  curves,  the  partial  differ- 
entials are  denned  as  : 

dD/8  log  i  =  7»  (intensity-scale  gamma)  (103} 

D  D/D  log  t  ==  yt  (time-scale  gamma)  (104} 
Hence  102  may  be  rewritten  in  simplified  form  : 


The  slope  of  the  reciprocity  failure  curve  equals  one  minus  the  ratio 
of  intensity-scale  gamma  to  time-scale  gamma.  At  high  intensities, 
such  as  used  in  sound-film  recording,  this  slope  is  positive;  hence 
the  intensity-scale  gamma  is  smaller  than  the  time-  scale  gamma. 

SYMBOLS 

Symbol  Definition  Dimension 

c     =  **  =  t*Ti 

D        =  Density 

D        =  Saturation  density 

e         =  it  =  relative  exposure 

/         =  1/7/  =  fading  constant  sec."1 

i*x 
F(x)    =  I    (\  —  e~x}  dx/x  =  integral  function 

Jo 

g         =  fraction  of  activated  grains  in  an  emulsion  layer 
i         =  l/r»  =  relative  intensity  sec."1 

p        =  probability  factor 

2    Cx 

Px      =  —  —  I     €—x*dx  =  probability  function 

^/TfJO 

r  =  fraction  of  exposed  grains  in  an  emulsion  layer 

R  =  fraction  of  exposed  grains  in  the  entire  emulsion 

t  =  exposure  time  sec. 

tb  =  blocking  time  sec. 

/2  =  electron  time  interval  for  second  step  sec. 

Tf  =  fading  time  sec. 

Ti  =  electron  time  interval  for  first  step  sec. 

u  =  absorption  constant  cm.  -1 

y  =  extension  into  depth  of  emulsion  cm. 

Y  =  total  thickness  of  emulsion  cm. 

a,  b,  k,  j,  m,  n,  q,  U,  V,  x,  z  =  auxiliary  symbols,   explained  in 

appendix 

e  =  basis  of  natural  logarithms 

r  =  transparency  of  emulsion 

6  =  partial  differential 

DISCUSSION 

MR.  ALBURGER:  What  is  the  mechanism  by  which  the  electrons  deposit  the 
silver  atoms  on  the  crystal  surface? 

MR.  ALBERSHEIM:  I  am  not  a  physicist;  I  can  only  guess  that  every  atom  in  the 
molecular  or  crystal  array  has  a  certain  electrical  field  surrounding  it,  and  we 


Jan.,  1939]  LATENT  IMAGE  THEORY  103 

know  that  unless  the  electron  has  a  certain  speed  it  will  be  captured.  A  sharp 
corner  produces  a  strong  electric  field;  in  ordinary  wire  we  have  corona  effects 
that  we  would  not  have  if  the  wire  were  round  and  polished.  Something  like 
that  happens  in  the  crystal,  and  at  points  of  physical  or  chemical  irregularity  the 
electric  field  becomes  strong  enough  to  capture  and  retain  the  electron.  When  it 
is  captured  the  excess  electron  will  be  attracted  to  the  silver  and  rip  it  loose  from 
the  bond  to  the  adjacent  chlorine  atom,  and  deposit  it  no  longer  as  an  ion  but 
as  a  silver  atom.  The  bromide  shifts,  and  if  it  finds  a  resting  place  in  the  adjacent 
gelatin  matrix  it  will  probably  come  to  rest  there.  * 

MR.  GOLDSMITH:  If  a  fully  exposed  negative  is  wound  up  on  a  reel  over  a  nega- 
tive of  a  much  less  brightly  illuminated  subject,  and  the  entire  negative  is  pre- 
served some  little  time  before  development,  is  there  found  to  be  any  trace  of  image 
transfer  due  to  light  re-emission  from  one  end  of  the  film  to  the  other? 

MR.  ALBERSHEIM  :  I  have  not  found  it  in  the  literature,  but  by  word  of  mouth  it 
has  been  reported  to  me  that  such  is  the  case,  that  under  some  conditions  a  pic- 
ture can  be  transferred  from  one  emulsion  to  an  adjacent  one.  It  may  be  that 
this  effect  is  not  as  dangerous  as  it  seems,  because  the  re-emitted  electrons  have 
a  threshold  value  and  it  is  possible  that  it  might  be  easier  to  obtain  the  effect  if 
the  second  emulsion  has  been  sensitized  to  infrared. 

MR.  DAVIS:  I  think  the  transfer  of  the  picture  from  one  film  to  another  is  the 
work  of  Sir  William  Abney,  of  England,  some  years  ago,  who  made  this  experi- 
ment. He  coated  a  plate  and  exposed  it,  and  coated  another  emulsion  on  top ;  then 
after  developing,  he  stripped  the  coat  and  found  a  picture  on  the  second  coating. 
This  probably  was  a  development  effect  and  not  a  bona  fide  latent  image  transfer. 

MR.  FRAYNE:  In  view  of  the  fact  that  the  energy  of  the  bullets  you  speak  of  is 
directly  proportional  to  the  frequency  of  the  light,  is  the  shape  of  the  catenary 
curves  dependent  upon  the  frequency  of  the  light  that  is  used? 

MR.  ALBERSHEIM:  The  influence  of  color  is  surprisingly  small.  I  thought  there 
would  be  such  an  effect,  but  this  was  disproved  by  the  Kodak  Research  Labora- 
tories. Webb  made  an  investigation  of  the  influence  of  color  from  green,  yellow, 
and  reddish  light,  so  far  as  the  emulsion  would  take  to  ultraviolet  of  3600  A, 
and  the  curves,  while  they  came  to  different  densities,  were  surprisingly  close  to 
parallelism.  The  curve  seems  just  to  shift  in  intensity,  but  not  in  character. 

MR.  FRAYNE:  Then  in  this  case  how  do  you  explain  the  change  of  gamma  with 
wavelength? 

MR.  ALBERSHEIM:  That  is  probably  due  to  a  resonance  effect.  The  emulsion 
is  most  sensitive  to  a  certain  wavelength,  from  4000  to  4400  A.  So  long  as  the 
impinging  light  falls  within  that  range  nearly  every  grain  will  be  exposed,  and 
high  ultimate  densities  and  high  gammas  result.  If  we  go  down  to  3500  A, 
many  grains  will  not  be  exposed  at  all,  so  the  emulsion  acts  as  if  it  had  fewer 
grains.  We  get  lower  gamma  and  lower  ultimate  density,  and  a  coarser  grained 
picture,  relatively  speaking.  The  grains  are  as  big  but  there  are  fewer  available. 

MR.  GOLDSMITH:  Would  you  assume  that  the  re-emitted  light  is  of  the  same 
wavelength  as  the  original  light  producing  the  image?  In  other  words,  if  a  colored 
object  is  photographed,  would  a  colored  image  be  released? 

*  Experiments  conducted  after  the  date  of  this  discussion  showed  trace  of  this 
effect. 


104  W.  J.  ALBERSHEIM 

MR.  ALBERSHEIM:  I  believe  not.  No  matter  how  hard  the  light  the  electron 
will  probably  lose  some  energy,  and  when  it  is  finally  absorbed  it  is  absorbed 
with  just  enough  energy  to  liberate  the  silver  atom.  The  excess  is  dissipated,  then, 
perhaps  as  light  or  heat,  so  when  the  electron  is  re-emitted  it  will  probably  be  re- 
emitted  nearly  monochromatically. 

MR.  JONES:  At  the  ultraviolet  end,  at  least,  the  exposing  radiation  is  absorbed 
by  the  thin  layers,  so  you  are  working  with  a  thinner  and  thinner  emulsion,  which 
of  course  gives  a  lower  gamma.  That  is  purely  absorption.  You  do  not  have 
to  invoke  resonance  to  explain  that. 

MR.  ALBERSHEIM:  If  the  emulsion  were  exposed  to  ultraviolet  of  sufficient 
density  you  would  finally  penetrate  through,  so  that  the  ultimate  density  at  very 
high  exposures  would  still  be  the  same.  It  would  take  very  much  more  light  to 
penetrate  to  the  bottom,  and  I  believe  there  must  be  some  other  effect  involved 
because  the  ultimate  gamma  is  lower.  Also,  if  you  expose  with  red  light  or  green- 
ish light,  which  penetrates  all  the  way  through  the  emulsion,  you  still  get  the 
lower  gamma.  There  may  be  a  superposition  of  two  effects  there. 

MR.  SANDVIK:  If  the  change  of  gamma  with  wavelength  is  a  resonance  phe- 
nomenon, how  would  you  explain  the  change  in  gamma  when  you  use  yellow  dye 
in  an  emulsion  and  use  blue  light? 

MR.  ALBERSHEIM:  That  would  probably  be  due  to  the  absorption  effect  that 
Dr.  Jones  has  mentioned.  However,  the  resonance  effect  is  present  because  there 
is  a  fairly  narrow  absorption  band  in  most  emulsions.  A  single  silver  halide 
crystal  exposed  to  light  has  a  resonance  frequency  at  about  4000  A  where  it  ab- 
sorbs the  maximum  number  of  photons  and  is  photographically  most  effective. 

MR.  SANDVIK:  It  does  that  no  matter  what  wavelength  radiation  is  lost. 
The  gamma  is  greatly  decreased  to  the  extent  that  the  absorption  takes  place 
with  the  wavelength  that  is  used. 

MR.  ALBERSHEIM:  I  am  glad  to  hear  so.  In  that  case  we  do  not  need  ultra- 
violet for  the  sound-films  to  obtain  reduced  gamma.  We  can  use  the  yellow  dye, 
which  we  have  been  preferring  all  along. 


THE  EVALUATION  OF  MOTION  PICTURE  FILMS 
BY  SEMIMICRO  TESTING* 


J.  E.  GIBSON**  AND  C.  G.  WEBERf 

Summary. — Test  methods  for  the  evaluation  of  motion-picture  film  for  permanent 
records  require  test  specimens  too  large  to  be  removed  from  certain  archival  films. 
To  assist  those  charged  with  the  preservation  of  such  films  in  determining  the  quality 
and  checking  the  condition  of  them,  suitable  semimicro  methods  were  developed  for 
acidity,  viscosity,  and  residual  hypo  content.  Specimens  as  small  as  7  mg.  in  weight, 
removed  from  the  film  with  a  small  hand  punch,  gave  satisfactory  results  for  the 
purpose. 

(I)  Introduction 

(II)  Experimental  testing 
(1)  Acidity 

(2}  Specific  viscosity 
(3)  Residual  hypo 

(III)  Summary  and  conclusions 

(I)     INTRODUCTION 

Certain  repositories,  such  as  The  National  Archives  and  some  film 
libraries,  are  called  upon  to  preserve  films  which  can  not  be  tested 
by  the  methods  usually  recommended  for  the  evaluation  of  film  for 
permanent  records.  From  these  films  it  is  not  possible  to  obtain 
test  specimens  of  sufficient  size  for  the  usual  tests  without  destroying 
some  of  the  photographic  images  or  making  the  film  unserviceable 
otherwise.  It  is  important  that  the  condition  of  such  films  be  de- 
termined. The  nitrate  films  are  chemically  unstable,  and  successful 
preservation  of  records  contained  on  them  requires  that  they  be  ex- 
amined periodically  so  that  disintegration  can  be  anticipated  and 
duplicates  made  of  the  records  before  they  are  impaired  by  visible 
deterioration.  Good  acetate  film  is  stable,  but  it  should  be  tested 
before  placing  it  in  storage  to  find  if  it  was  properly  made  and  proc- 
essed. Also,  subsequent  testing  of  it  may  be  desirable,  particularly 

*  Presented  at  the  1938  Fall  Meeting  at  Detroit,  Mich. ;     received  October   3, 
1938. 

**  The  National  Archives,  Washington,  D.  C. 
t  National  Bureau  of  Standards,  Washington,  D.  C. 

105 


106  J.  E.  GIBSON  AND  C.  G.  WEBER  [j.  s.  M.  P.  E. 

if  it  should  be  exposed  to  unfavorable  storage  conditions.  The 
semimicro  methods  described  in  this  article  were  developed  to  permit 
the  testing  of  such  films  without  removing  test  specimens  large  enough 
to  impair  the  film  as  regards  legibility  and  serviceability. 

(II)     EXPERIMENTAL  TESTING 

In  studies1  of  the  stability  of  photographic  films,  tests  for  copper 
number,  viscosity,  acidity,  residual  hypo  (sodium  thiosulfate),  and 
flexibility  were  found  to  be  of  value.  These  tests  were  recommended2 
for  the  evaluation  of  film  for  permanent  records.  Hence,  the  micro 
methods  developed  were  modifications  of  some  of  these  methods. 
The  value  of  each  of  the  proposed  methods  was  determined  by  testing 
films  that  had  been  subjected  to  accelerated  aging  at  100°C  in  oven- 
dry  air  for  various  periods  of  time.  The  data  are  thus  comparable 
with  those  obtained  by  Hill  and  Weber1  with  test  specimens  of  normal 
amount.  The  micro  tests  were  made  with  test  specimens  weighing 
only  7  mg.  each,  which  were  removed  from  the  films  with  a  J /4-inch 
hand  punch  without  causing  appreciable  damage  to  the  film.  The 
value  of  the  micro  test  was  judged  by  comparing  the  results  of  them 
with  results  obtained  with  the  usual  methods.  The  micro  methods 
developed  were  for  acidity,  viscosity,  and  residual  hypo. 

(1)  Acidity. — The  acidity  of  the  film  was  determined  in  the  follow- 
ing manner.     A  single  punching  (wt.  0.007  g.)  of  film,  including  both 
base  and  emulsion,  was  transferred  to  a  test  tube  and  5  ml.  of  acetone, 
containing  10  per  cent  of  water  by  volume,  was  added.     After  com- 
plete dispersion  of  the  film  base,  the  acidity  in  £H  units  was  deter- 
mined by  means  of  a  commercial  micro  £H-meter.     The  results  are 
shown  in  Fig.  1  in  comparison  with  ^>H  values  obtained  with  the  same 
apparatus  for  seven  punchings  of  film  weighing  0.049  gram  in  5  ml.  of 
acetone  containing  10  per  cent  of  water  by  volume. 

The  water  and  acetone  were  purified  by  distillation  and  the  com- 
bined solvent  had  a  pH  of  7  ^  0.4.  Duplicate  determinations  on  the 
film  agreed  within  0.1  pH  unit. 

(2)  Specific    Viscosity. — A  punching  of  film   (wt.   0.007  g.)   was 
transferred  to  a  test  tube  and  dissolved  in  5  ml.  of  acetone  measured 
at  30°  =*=  0.02°C.     After  solution  of  the  film  base  was  complete  and 
the  mixture  homogeneous,  3  ml.  of  the  solution  was  transferred  to  an 
Ostwald  viscosity  pipette  immersed  in  a  constant-temperature  bath 
(30°  ±  0.02°C),  and  allowed  to  stand  until  temperature  equilibrium 
was  reached.     The  time  of  flow  of  the  solution  through  the  capillary 


Jan.,  1939]          EVALUATING  FILMS  BY  SEMIMICRO  TESTING 


107 


of  the  pipette  was  measured  with  a  stop-watch  which  could  be  read 
to  one-fifth  second.  The  time  of  flow  of  the  pure  solvent  was  also 
measured.  Not  less  than  three  or  four  determinations  were  made  for 
each  solution,  the  values  agreeing  within  two-  or  three-tenths  of  a 
second.  The  relative  viscosity  was  then  calculated  as  the  ratio  of 
the  time  of  flow  of  the  solution  to  the  time  of  flow  of  the  solvent. 

Fig.  2  shows  the  results  of  these  measurements  compared  with  re- 
sults obtained  by  Hill  and  Weber  with  one-gram  samples.  The 
different  scales  were  necessary  because  different  values  are  obtained 
by  the  two  methods. 


USUAL   METHOD  -7  PUNCHINGS  (.049  6M  > 

Q   ACETATE    FILM 

•   NITRATE    FILM 

SEMI-MICRO   METHOD -I  PUNCHING    (.007  GM)' 

§   ACETATE     FILM 
NITRATE     FILM 


10  20 

TIME  OF   HEATING-  DAYS 


FIG.  1.  Effects  of  accelerated  aging  on  pH  of  cellulose 
acetate  and  cellulose  nitrate  films ;  results  obtained  by  the 
semimicro  method  compared  with  those  obtained  on  larger 
test  specimens. 


(3)  Residual  Hypo. — The  method  used  for  detecting  the  presence  of 
residual  hypo  (sodium  thiosulfate)  in  films  is  a  modification  of  the 
test  proposed  by  Crab  tree  and  Ross.3  The  method  as  modified  con- 
sists in  placing  a  single  punching  of  film  on  a  glass  slide,  adding  two 
drops  of  mercuric  chloride  test  solution  to  the  specimen  in  such  a 
manner  that  the  solution  flows  over  the  specimen  and  onto  the  glass, 
and  observing  any  turbidity  that  develops  in  the  solution.  The  test 
solution  contains  25  grams  of  mercuric  chloride  and  25  grams  of 
potassium  bromide  in  a  liter  of  aqueous  solution.  The  film  is  placed 
on  the  glass  with  the  emulsion  side  up,  and  is  allowed  to  stand  for  2 
or  3  minutes  after  the  addition  of  the  test  solution.  It  was  found- 


108 


J.  E.  GIBSON  AND  C.  G.  WEBER          [j.  s.  M.  P.  E. 


that  any  turbidity  of  the  solution  can  be  best  detected  with  the  un- 
aided eye  when  the  glass  is  held  in  the  light  so  that  the  angle  of  in- 
cidence is  approximately  90  degrees. 

If  sodium  thiosulfate  is  present,  it  reduces  mercuric  ion,  and  an  in- 
soluble mercurous  compound  is  formed  which  causes  turbidity.  If 
no  thiosulfate  is  present,  the  solution  on  the  glass  remains  clear,  al- 
though the  silver  image  is  bleached  white.  Positive  tests  were  ob- 
tained in  this  manner  on  single  punchings  taken  from  film  that  con- 
tained less  than  0.05  mg.  of  hypo  per  square-inch.  Positive  tests 


•WAI  METHOD 
ACETATE  FILM 
NITRATE  FILM 
MI-MICRO  METHOD 

§  ACETATE  FILM 
NITRATE  FILM 


TIME    OF    HEATING -DAYS 


FIG.  2.  Effects  of  oven  aging  on  viscosity  of  cellulose 
acetate  and  cellulose  nitrate  films  as  determined  by  the 
semimicro  methods  and  by  the  usual  methods. 

were  also  obtained  with  solutions  containing  10  parts  of  hypo  per 
million  parts  of  water  when  a  large  drop  of  the  solution  was  added  to  a 
drop  of  the  test  solution. 

(ni)     SUMMARY  AND  CONCLUSIONS 

Motion  picture  films  that  can  not  be  sampled  for  testing  by  the 
usual  methods  can  be  tested  by  the  semimicro  methods.  The  ap- 
proximate quality  and  condition  of  the  film  can  be  determined  by 
tests  for  acidity,  viscosity,  and  residual  hypo,  using  specimens  weigh- 
ing only  7  mg.  each,  which  can  be  taken  from  the  film  by  means  of  a 
small  hand  punch  without  appreciable  damage  to  the  film.  These 
methods  are  not  recommended  for  use  in  selecting  permanent  record 
film.  However,  they  are  recommended  to  archivists  and  librarians 
for  determining  the  condition  of  finished  films  given  them  for  custody. 


Jan.,  1939  ]          EVALUATING  FlLMS  BY  SEMIMICRO  TESTING  109 

With  these  tests,  the  approximate  condition  of  the  films  can  be  found, 
and  the  necessity  of  making  duplicate  copies  can  be  determined  be- 
fore the  damage  to  the  films  by  deterioration  is  serious  enough  to  be 
visible.  The  values  obtained  by  the  semimicro  methods  will  differ 
somewhat  from  those  obtained  by  the  usual  methods,  but  they  ap- 
pear to  show  the  extent  of  deterioration  under  accelerated  aging 
equally  well.  When  absolute  values  are  used  in  judging  the  condi- 
tion of  a  film  in  question,  those  obtained  by  the  semimicro  methods 
should  be  compared  with  values  obtained  for  new  film  by  these 
methods. 

REFERENCES 

1  /.  Research,  Nat.  Bur.  Standards,  17  (1936),  p.  871;    RP950. 

*  Muse.  Pub.  Nat.  Bur.  Standards,  M158  (1937). 

3  CRABTREE,  J.  I.,  AND  Ross,  J.  F.:  "A  Method  of  Testing  for  the  Presence  of 
Sodium  Thiosulfate  in  Motion  Picture  Films,"  J.  Soc.  Mot.  Pict.  Eng.,  XIV 
(April,  1930),  No.  4,  p.  419. 

DISCUSSION 

MR.  CRABTREE:  Our  recent  researches  have  indicated  that  the  milky  compound 
formed  by  reaction  of  the  mercuric  chloride  with  hypo  is  not  mercurous  chloride 
but,  rather,  a  double  compound  of  mercuric  sulfide  and  mercuric  chloride  having 
the  formula  2HgS  HgCl2.  Such  a  compound  has  been  described  by  H.  Rose 
(Poggendorff,  Annalen  der  Physik,  13:  59,  1828)  and  by  Th.  Poleck  and  C.  Goercki 
(Berichte,  21:  2412-2417,  1888). 


CURRENT   LITERATURE    OF   INTEREST   TO   THE   MOTION   PICTURE 

ENGINEER 

The  editors  present  for  convenient  reference  a  list  of  articles  dealing  with  subjects 
cognate  to  motion  picture  engineering  published  in  a  number  of  selected  journals. 
Photostatic  copies  may  be  obtained  from  the  Library  of  Congress,  Washington,  D.  C., 
or  from  the  New  York  Public  Library,  New  York,  N.  Y.  Micro  copies  of  articles 
in  magazines  that  are  available  may  be  obtained  from  the  Bibliofilm  Service,  Depart- 
ment of  Agriculture,  Washington,  D.  C. 


Journal  of  the  Acoustical  Society  of  America 

10  (Oct.,  1938),  No.  2 

Absorption  of  Sound  in  Carbon  Dioxide  and  Other  Gases 
(pp.  89-97) 

Measurement  of  Absorption  in  Rooms  with  Sound  Ab- 
sorbing Ceilings  (pp.  98-101) 

Absorption  Effects  in  Sound  Transmission  Measure- 
ments (pp.  102-104) 

Absolute  Sound  Measurements  in  Liquids  (pp.  105-111) 
Theory  of  the  Chromatic  Stroboscope  (pp.  112-118) 

Adjustable  Tuning  Fork  Frequency  Standard  (pp.  119- 

127) 
Recent  Advances  in  the  Use  of  Acoustic  Instruments  for 

Routine  Production  Testing  (pp.  128-134) 
Frequency  Ratios  of  the  Tempered  Scale  (pp.  135-136) 
Harmonic  Structure  of  Vowels  in  Singing  in  Relation  to 

Pitch  and  Intensity  (pp.  137-146) 
Apparatus  for  Direct-Recording  the  Pitch  and  Intensity 

of  Sound  (pp.  147-149) 

American  Cinematographer 

19  (Oct.,  1938),  No.  10 

Flashes  Across  Nearly  Sixty  Years  (pp.  403-404) 
Dunning  Has  Three-Color  Process  Now  Ready  to  Go 

(pp.  406,  416) 
Mole-Richardson   Introduces   Duarc,   New  Automatic 

Broadside  (pp.  407,  416) 
Ingenious  Accessories  Simplify  Making  of  Special  Effects 

Shots  (pp.  408,  410) 

100  Watter  Throws  150 — and  Whiter  (p.  411) 
American  Cameramen  Lead.  .  .  Pasternak  (pp.  412-414) 

110 


V.  O.  KNUDSEN  AND 
E.  F.  FRICKE 

J.  R.  POWER 

P.  E.  SABINE  AND 
L.  G.  RAMER 
E.  KLEIN 
R.  W.  YOUNG  AND 

A.  LOOMIS 

O.  H.  SCHUCK 

B.  FOULDS 

C.  WILLIAMSON 

B.  STOUT 

J.  OB  AT  A  AND 
R.  KOBAYASHI 


J.  GEVAERT 


G.  TEAGUE 
J.  PASTERNAK 


CURRENT  LITERATURE 


111 


American  Cinematographer 

19  (Nov.,  1938),  No.  11 

What's  Wrong  with  Cinematography?     (pp.  449,  457) 
Reeves  Single  System  Sound  Fits  Any  Camera  (pp.  454- 

455) 

New  Berndt-Maurer  Sound  Tract  (pp.  456) 
20th-Fox  Installs  New  Make-Up  Lamps  (p.  479) 

British  Journal  of  Photography 

85  (Sept.  9,  1938),  No.  4088 
Aluminum  as  a  Photographic  Base  (pp.  568-570) 

Journal  of  the  British  Kinematograph  Society 

1  (Oct.,  1938),  No.  3 

Modern  Electric  Discharge  Lamps  and  Their  Applica- 
tion to  Kinematography  (pp.  158-174) 
Volume  Range  Expanders  (pp.  175-187) 
Manufacture  of  Motion  Picture  Film  (pp.  188-204) 
An  Optical  System  for  Sound  Reproduction  (pp.  209- 
213) 

Bulletin  de  la  Societe  Francaise  de  Photographic  et  de 
Cinematographic 

25  Sere.  3  (Sept.,  1938),  No.  9 

Sur  L'Obtention  de  Negatifs  Photographiques  a  Grains 
Fins  a  Partir  d'Emulsions  ou  d'Images  a  Gros  Grains. 
(Obtaining  Fine  Grain  Negatives  Starting  with  Large 
Grain  Emulsions  or  Images)  (pp.  145-150) 

Educational  Screen 

17  (Sept.,  1938),  No.  7 

Motion  Pictures — Not  for  Theaters,  Pt.  I  (pp.  211- 
215) 

17  (Oct.,  1938),  No.  8 
Motion  Pictures — Not  for  Theaters.     Pt.  II  (pp.  249- 

253) 
Preparing  Sound  Film  Strips  (pp.  254-256) 

Electronics 

11  (Oct.,  1938),  No.  10 

A  Laboratory  Television  Receiver — IV  (pp.  16-19) 
A  Shielded  Loop  for  Noise  Reduction  in  Broadcast  Re- 
ception (pp.  20-22) 
Squeeze  or  Matted  Track  (p.  23) 
An  Electric  Timing  Device  (pp.  28-29) 

Ideal  Kinema 

6  (Oct.  13,  1938),  No.  71 

The  Stableford  All-Metal  Screen,  How  It  Is  Constructed 
(P-  33) 


H.  W.  GREENWOOD 


W.  R.  STEVENS 
J.  Mom 
A.  E.  AMOR 

J.  H.  McLsoD  AND 
F.  E.  ALTMAN 


A.  SEYEWETZ 


A.  E.  KROWS 


A.  E.  KROWS 
C.  R.  THOMAS 


D.  G.  FINK 

S.  GOLDMAN 
J.  K.  HILLIARD 
R.  W.  CARLSON 


112 


CURRENT  LITERATURE 


Kinematographic  Weekly 

259  (Sept.  29,  1938),  No.  1641 
Metals  as  Base  for  Picture  Film  (p.  36) 

260  (Oct.  27,  1938),  No.  1645 
Metal  Film  Projection  (p.  41) 

Television  Principles  in  Kinematography  (p.  41) 

International  Photographer 

10  (Oct.,  1938),  No.  9 
Duplex  Production  Printer  (pp.  6-7) 
Ultra-Fidelity  Recorder  (p.  7) 
Gevaert  Revives  35-Mm.  Raw  Stock  (p.  9) 
Technicolor  Expands  (p.  9) 
Projection-Revision  of  SMPE  Standards  (pp.  24-27) 

Kinotechnik 

20  (Oct.,  1938),  No.  10 
Sicherheitsnlm  im  Normalformat  (35  Mm.  Safety  Film) 

(pp.  255-256) 
Bildfilm  und  Magnetton  (Magnetic  Sound  Recording  on 

Film)  (pp.  256-257) 

Methoden  zur  Messung  des  photographischen  Gleich- 
richtereffektes  (Method  of  Measuring  the  Photo- 
graphic Rectifying  Effect)  (pp.  258-264) 

Zwei  neue  Aufnahme-Materialen ;  Agfa  Superpanfilm 
und  Agfa  Ultrarapidfilm  (Two  New  Films;  Agfa 
Superpan  and  Agfa  High-Speed  Film)  (pp.  264-267) 

Moderne  Wiedergabegerate  fur  16-Mm.-Tonfilm  (Mod- 
era  16-Mm.  Sound-Film  Projectors)  (pp.  268-269) 

International  Projectionist 

13  (Oct.,  1938),  No.  10 
Advance   Preparations   Minimize   Sound    Emergencies 

(pp.  7-10) 

Television  and  Its  Effect  Upon  the  Motion  Picture 
Theatre)  (pp.  12-14) 

A  Higher-Efficiency  Condensing  System  for  Tungsten- 
Filament  Projectors  (pp.  14-16) 

Projection  Possibilities  of  Mercury  Vapor  Discharge 
Lamp  (pp.  17-18) 

Photographische  Korrespondenz 

74  (Oct.,  1938),  No.  10 

Fortschritte  der  Kinematographie  im  Jahre  1937  (Prog- 
ress of  Photography  in  1937)  (pp.  164-167) 


C.  N.  BATSEL 


J.  GRASSMAN 


H.  PETERSEN 


A.  NARATH  AND 
W.  Vox 


A.  SCHILLING 


WEINBERGER 


A.  NADELL 

F.  WALDROP  AND 
J.  BORKIN 

F.  E.  CARLSON 
R.  H.  CRICKS 


H.  PANDER 


SPRING,  1939,  CONVENTION 
SOCIETY  OF  MOTION  PICTURE  ENGINEERS 


HOLLYWOOD  ROOSEVELT  HOTEL 
HOLLYWOOD,   CALIFORNIA 
APRIL  17th-21st,  INCLUSIVE 


Officers  and  Committees  in  Charge 

E.  A.  WILLIFORD,  President 

N.  LEVINSON,  Executive  V ice-President 

W.  C.  KUNZMANN,  Convention  Vice-President 

J.  I.  CRABTREE,  Editorial  Vice-President 

L.  RYDER,  Chairman,  Pacific  Coast  Section 

H.  G.  TASKER,  Chairman,  Local  Arrangements  Committee 

J.  HABER,  Chairman,  Publicity  Committee 

Pacific  Coast  Papers  Committee 

L.  A.  AICHOLTZ,  Chairman 

O.  O.  CECCARINI  W.  A.  MUELLER 

G.  CHAMBERS  H.  G.  TASKER 

L.  D.  GRIGNON  W.  H.  ROBINSON,  JR. 

Reception  and  Local  Arrangements 

H.  G.  TASKER,  Chairman 

N.  LEVINSON  G.  F.  RACKETT  E.  HUSE 

K.  F.  MORGAN  H.  W.  MOYSE  L.  L.  RYDER 

P.  MOLE  W.  MILLER  J.  O.  AALBERG 

A.  M.  GUNDELFINGER       J.  A.  BALL  R.  H.  McCULLOUGH 

H.  W.  REMERSCHIED       W.  A.  MUELLER  J.  M.  NICKOLAUS 

G.  S.  MITCHELL  C.  L.  LOOTENS  E.  H.  HANSEN 

Registration  and  Information 

W.  C.  KUNZMANN,  Chairman 

S.  HARRIS  C.  W.  HANDLEY 

E.  R.  GEIB  W.  R.  GREENE 

Hotel  and  Transportation 

G.  A.  CHAMBERS,  Chairman 

W.  C.  HARCUS  C.  DUNNING  W.  E.  THEISEN 

G.  HOUGH  E.  C.  RICHARDSON  D.  P.  LOYE 

B.  KREUZER  K.  STRUSS  O.  O.  CECCARINI 
H.  W.  REMERSCHIED       J.  C.  BROWN                          C.  J.  SPAIN 

113 


114  SPRING,  1939,  CONVENTION  [j.  s.  M.  p.  E. 

Convention  Projection 

H.  GRIFFIN,  Chairman 

J.  O.  AALBERG  H.  A.  STARKE  J.  DURST 

C.  W.  HANDLEY  L.  E.  CLARK  R.  H.  McCuLLOUGH 

W.  F.  RUDOLPH  M.  S.  LESHING  H.  C.  SILENT 

J.  M.  NICKOLAUS  A.  F.  EDOUART  H.  I.  REISKIND 

J.  K.  HILLIARD  I.  SERRURIER  W.  W.  LINDSAY,  JR. 

Officers  and  Members  of  Los  Angeles  Projectionists  Local  No.  150 

Banquet  and  Dance 

N.  LEVINSON,  Chairman 

H.  T.  KALMUS  G.  S.  MITCHELL  G.  F.  RACKETT 

E.  HUSE  P.  MOLE  J.  O.  AALBERG 

L.  L.  RYDER  H.  G.  TASKER  K.  F.  MORGAN 

C.  DUNNING  W.  MILLER  H.  W.  MOYSE 

G.  A.  MITCHELL  R.  H.  MCCULLOUGH  J.  L.  COURCIER 

Ladies'  Reception  Committee 

MRS.  N.  LEVINSON,  Hostess 

assisted  by 

MRS.  E.  C.  RICHARDSON  MRS.  P.  MOLE  MRS.  E.  HUSE 

MRS.  G.  F.  RACKETT       MRS.  C.  W.  HANDLEY  MRS.  L.  L.  RYDER 

MRS.  H.  W.  MOYSE         MRS.  K.  F.  MORGAN  MRS.  J.  O.  AALBERG 

MRS.  H.  G.  TASKER        MRS.  W.  MILLER  MRS.  R.  H.  MCCULLOUGH 

MRS.  L.  E.  CLARK  MRS.  A.  M.  GUNDELFINGER  MRS.  C.  DUNNING 

Publicity 

J.  HABER,  Chairman 

L.  A.  AICHOLTZ  W.  R.  GREENE  S.  HARRIS 

W.  A.  MUELLER  G.  CHAMBERS  A.  M.  GUNDELFINGER 

Equipment  Exhibit 

J.  G.  FRAYNE,  Chairman 

P.  MOLE  C.  R.  DAILY  H.  W.  REMERSCHIED 

J.  DURST  S.  HARRIS  C.  N.  BATSEL 

Headquarters 

Headquarters  of  the  Convention  will  be  the  Hollywood-Roosevelt  Hotel,  where 
excellent  accommodations  are  assured.  A  reception  suite  will  be  provided  for  the 
Ladies'  Committee,  and  an  excellent  program  of  entertainment  will  be  arranged 
for  the  ladies  who  attend  the  Convention. 

Special  hotel  rates,  guaranteed  to  SMPE  delegates,  European  plan,  will  be  as 
follows : 

One  person,  room  and  bath  $  3 . 50 

Two  persons,  double  bed  and  bath  5 . 00 

Two  persons,  twin  beds  and  bath  6 . 00 

Parlor  suite  and  bath,  1  person  8 . 00 

Parlor  suite  and  bath,  2  persons  12 . 00 


Jan.,  1939]  SPRING,  1939,  CONVENTION  115 

Indoor  and  outdoor  garage  facilities  adjacent  to  the  Hotel  will  be  available 
to  those  who  motor  to  the  Convention. 

Members  and  guests  of  the  Society  will  be  expected  to  register  immediately 
upon  arriving  at  the  Hotel.  Convention  badges  and  identification  cards  will 
be  supplied  which  will  be  required  for  admittance  to  the  various  sessions,  the 
studios,  and  several  Hollywood  motion  picture  theaters. 

Railroad  Fares 

The  following  table  lists  the  railroad  fares  and  Pullman  charges: 

Railroad 

Fare  Pullman 

City  (round  trip)  (one  way) 

Washington  $132.20  $22.35 

Chicago  90.30  16.55 

Boston  147.50  23.65 

Detroit  106.75  19.20 

New  York  139.75  22.85 

Rochester  124.05  20.50 

Cleveland  110.00  19.20 

Philadelphia  135.50  22.35 

Pittsburgh  117.40  19.70 

The  railroad  fares  given  above  are  for  round  trips,  sixty-day  limits.  Arrange- 
ments may  be  made  with  the  railroads  to  take  different  routes  going  and  coming, 
if  so  desired,  but  once  the  choice  is  made  it  must  be  adhered  to,  as  changes  in  the 
itinerary  may  be  effected  only  with  considerable  difficulty  and  formality.  Dele- 
gates should  consult  their  local  passenger  agents  as  to  schedules,  rates,  and  stop- 
over privileges. 

Technical  Sessions 

The  Hollywood  meeting  always  offers  our  membership  an  opportunity  to  be- 
come better  acquainted  with  the  studio  technicians  and  production  problems,  and 
arrangements  will  be  made  to  visit  several  of  the  studios.  The  Local  Papers 
Committee  under  the  chairmanship  of  Mr.  L.  A.  Aicholtz  is  collaborating  closely 
with  the  General  Papers  Committee  in  arranging  the  details  of  the  program. 
Complete  details  of  the  program  will  be  published  in  a  later  issue  of  the  JOURNAL. 

Semi- Annual  Banquet  and  Dance 

The  Semi- Annual  Banquet  of  the  Society  will  be  held  at  the  Hotel  on  Thursday, 
April  20th.  Addresses  will  be  delivered  by  prominent  members  of  the  industry, 
followed  by  dancing  and  entertainment.  Tables  reserved  for  8,  10,  or  12  persons; 
tickets  obtainable  at  the  registration  desk. 

Equipment  Exhibit 

An  exhibit  of  newly  developed  motion  picture  equipment  will  be  held  in  the 
Bombay  and  Singapore  Rooms  of  the  Hotel,  on  the  mezzanine.  Those  who  wish 
to  enter  their  equipment  in  this  exhibit  should  communicate  as  early  as  possible 
with  the  general  office  of  the  Society  at  the  Hotel  Pennsylvania,  New  York,  N.  Y. 


116  SPRING,  1939,  CONVENTION 

Motion  Pictures 

At  the  time  of  registering,  passes  will  be  issued  to  the  delegates  to  the  Conven- 
tion, admitting  them  to  the  following  motion  picture  theaters  in  Hollywood,  by 
courtesy  of  the  companies  named:  Grauman's  Chinese  and  Egyptian  Theaters 
(Fox  West  Coast  Theaters  Corp.),  Warner's  Hollywood  Theater  (Warner  Brothers 
Theaters,  Inc.),  Pantages  Hollywood  Theater  (Rodney  Pantages,  Inc.).  These 
passes  will  be  valid  for  the  duration  of  the  Convention. 

Inspection  Tours  and  Diversions 

Arrangements  are  under  way  to  visit  one  or  more  of  the  prominent  Hollywood 
studios,  and  passes  will  be  available  to  registered  members  to  several  Hollywood 
motion  picture  theaters.  Arrangements  may  be  made  for  golfing  and  for  special 
trips  to  points  of  interest  in  and  about  Hollywood. 

Ladies'  Program 

An  especially  attractive  program  for  the  ladies  attending  the  Convention  is 
being  arranged  by  Mrs.  N.  Levinson,  hostess,  and  the  Ladies'  Committee.  A 
suite  will  be  provided  in  the  Hotel,  where  the  ladies  will  register  and  meet  for 
the  various  events  upon  their  program.  Further  details  will  be  published  in  a 
succeeding  issue  of  the  JOURNAL. 

Points  of  Interest 

En  route:  Boulder  Dam,  Las  Vegas,  Nevada;   and  the  various  National  Parks. 

Hollywood  and  vicinity:  Beautiful  Catalina  Island;  Zeiss  Planetarium;  Mt. 
Wilson  Observatory;  Lookout  Point,  on  Lookout  Mountain;  Huntington  Li- 
brary and  Art  Gallery  (by  appointment  only) ;  Palm  Springs,  Calif. ;  Beaches  at 
Ocean  Park  and  Venice,  Calif.;  famous  old  Spanish  missions;  Los  Angeles  Mu- 
seum (housing  the  SMPE  motion  picture  exhibit);  Mexican  village  and  street, 
Los  Angeles. 

In  addition,  numerous  interesting  side  trips  may  be  made  to  various  points 
throughout  the  west,  both  by  railroad  and  bus.  Among  the  bus  trips  available 
are  those  to  Santa  Barbara,  Death  Valley,  Agua  Caliente,  Laguna,  Pasadena, 
and  Palm  Springs,  and  special  tours  may  be  made  throughout  the  Hollywood 
area,  visiting  the  motion  picture  and  radio  studios. 

On  February  18,  1939,  the  Golden  Gate  International  Exposition  will  open 
at  San  Francisco,  an  overnight  trip  from  Hollywood.  The  Exposition  will  last 
throughout  the  summer  so  that  opportunity  will  be  afforded  the  eastern  members 
to  take  in  this  attraction  on  their  convention  trip. 


SOCIETY  ANNOUNCEMENTS 

ELECTION  OF  SECTION  OFFICERS 

Results  of  the  election  of  officers  and  managers  of  the  Mid- West  and  Pacific 
Coast  Sections  of  the  Society  are  as  follows: 

(Mid-West} 

*S.  A.  LUKES,  Chairman 

C.  H.  STONE,  Past-Chairman  *J.  A.  DUBRAY,  Manager 

*G.  W.  BAKER,  Sec.-Treas.  **O.  B.  DEPUE,  Manager 

(Pacific  Coast) 
*L.  RYDEE,  Chairman 

J.  O.  AALBERG,  Past-Chairman  *C.  W.  HANDLEY,  Manager 

*A.  M.  GUNDELFINGER,  Sec.-Treas.  **W.  MILLER,  Manager 

*  Term  expires  December  31,  1939. 
**  Term  expires  December  31,  1940. 

Elections  of  officers  and  managers  of  the  Atlantic  Coast  Section  are  now  in 
progress  and  will  be  announced  in  the  next  issue  of  the  JOURNAL. 

ATLANTIC  COAST  SECTION 

At  a  meeting  of  the  Section  held  on  December  13th  at  the  studios  of  RCA 
Photophone,  Inc.,  New  York,  a  paper  was  presented  by  F.  C.  Gilbert  and  E.  S. 
Seeley  of  the  Altec  Service  Corporation,  New  York,  on  the  subject  of  "The 
Adjustable  Equalizer  as  a  Tool  for  Selecting  the  Best  Response  Characteristics." 
This  equalizer  is  a  device  that  can  be  inserted  into  theater  reproducing  systems 
for  determining  with  a  given  horn  system  what  characteristic  is  best  in  a  given 
house.  It  is  portable  and  can  be  carried  into  the  auditorium,  and  has  an  ex- 
tremely wide  range  of  variation. 

The  paper  was  presented  by  Mr.  Seeley  and  aroused  considerable  interest 
among  the  members  attending  the  meeting,  as  evidenced  by  the  protracted 
discussion  held  at  the  close  of  the  presentation.  A  demonstration  of  the  equalizer 
accompanied  the  presentation. 

MID-WEST   SECTION 

At  a  meeting  held  at  The  Western  Society  of  Engineers,  Chicago,  on  December 
6th,  Mr.  J.  Frankenberg  presented  a  paper  dealing  with  "Mechanical  Sound 
Recording  of  Film."  The  meeting  was  well  attended  and  the  presentation  was 
discussed  at  considerable  length. 

Announcement  of  the  officers  and  managers  of  the  Section  for  the  year  1939 
was  made  as  listed  above. 

117 


118  SOCIETY  ANNOUNCEMENTS 

PACIFIC  COAST  SECTION 

On  December  15th  a  meeting  of  the  Section  was  held  at  the  Walt  Disney 
Studios  in  Hollywood,  at  which  time  a  demonstration  of  the  recording  spectro- 
photometer  and  the  Disney  multiplane  camera  was  given  by  the  technical  staff 
of  the  Walt  Disney  Studios.  On  account  of  limited  accommodations,  the  meeting 
was  open  to  only  members  of  the  Society  and  was  well  attended.  The  presenta- 
tion elicited  much  interest  and  discussion. 

CONVENTION  ACKNOWLEDGMENTS 

Acknowledgment  is  due  to  many  companies  and  persons  for  their  cooperation 
in  arranging  and  conducting  the  Detroit  Convention,  held  on  October  31st- 
November  2nd,  with  headquarters  at  the  Hotel  Statler.  General  facilities  of 
the  Convention  were  arranged  by  Mr.  W.  C.  Kunzmann,  Convention  Vice-P resi- 
dent; Messrs.  H.  Griffin,  J.  Frank,  Jr.,  and  G.  Friedl,  Jr.,  in  charge  of  projection 
facilities;  Mr.  K.  Brenkert,  Chairman  of  the  Local  Arrangements  Committee; 
A.  J.  Bradford  and  J.  F.  Strickler  on  the  Local  Arrangements  Committee;  Mrs. 
J.  F.  Strickler,  hostess  in  charge  of  the  Ladies'  Committee;  Mr.  J.  Haber  and  F. 
Johntz  of  the  Publicity  Committee;  and  Mr.  E.  R.  Geib,  Chairman  of  the  Member- 
ship Committee. 

Credit  for  the  papers  program  and  technical  arrangements  are  due  to  Mr. 
J.  I.  Crabtree,  Editorial  Vice-P  resident,  and  Mr.  G.  E.  Matthews,  Chairman  of 
the  Papers  Committee. 

Among  the  companies  contributing  equipment  and  service  to  the  Convention 
were  the  following:  International  Projector  Corporation,  National  Carbon 
Company,  National  Theatre  Supply  Company,  Raven  Screen  Company,  East- 
man Kodak  Company,  Bausch  &  Lomb  Optical  Company,  RCA  Manufacturing 
Company,  Brenkert  Light  Projection  Corporation,  Jam  Handy  Pictures  Cor- 
poration, and  the  Detroit  Local  199  IATSE. 

The  Society  is  indebted  to  the  following  companies  for  the  films  loaned  for 
the  motion  picture  performance  held  on  the  evening  of  Monday,  October  31st: 
RKO  Radio  Pictures,  Paramount  Pictures,  Inc.,  Eastman  Kodak  Company, 
Technicolor  Motion  Picture  Corporation,  March  of  Time,  and  Walt  Disney 
Productions,  Ltd. 

Acknowledgment  is  due  also  to  the  United  Detroit  Theaters  Corporation  and 
the  Fox  Detroit  Theater  for  supplying  passes  to  members  and  guests  during  the 
week  of  the  Convention. 

ADMISSIONS  COMMITTEE 

The  following  applicants  have  been  admitted  by  vote  of  the  Board  of  Governors 
to  the  Active  grade: 

CASE,  P.  H.  HONAN,  E.  M. 

28  West  23rd  St.,  6601  Romaine  St., 

New  York,  N.  Y.  Los  Angeles,  Calif. 

FESSLER,  F.  D.  SAWYER,  C.  R. 

4431  West  Lake  St.,  6601  Romaine  St., 

Chicago,  111.  Los  Angeles,  Calif. 

WALKER,  H.  S. 

1620  Notre  Dame  St.  West, 
Montreal,  Canada 


JOURNAL 

OF  THE  SOCIETY  OF 

MOTION  PICTUTE  ENGINEERS 


Volume  XXXII  February,  1939 


CONTENTS 

Page 
Some  Television  Problems  from  the  Motion  Picture  Standpoint 

G.  L.  BEERS,  E.  W.  ENGSTROM,  AND  I.  G.  MALOFF     121 

Some  Production  Aspects  of  Binaural  Recording  for  Sound 

Motion  Pictures 

W.  H.  OFFENHAUSER,  JR.,  AND  J.  J.  ISRAEL     139 

Coordinating  Acoustics  and  Architecture  in  the  Design  of  the 
Motion  Picture  Theater.  .  C.  C.  POT  WIN  AND  B.  SCHL  ANGER  156 

Characteristics  of  Film  Reproducer  Systems 

F.  DURST  AND  E.  J.  SHORTT     169 

Some  Practical  Accessories  for  Motion  Picture  Recording 

R.  O.  STROCK     188 

The  Lighting  of  Motion  Picture  Theater  Auditoriums 

F.  M.  FALGE  AND  W.  D.  RIDDLE     201 

Revised  Standard  Electrical  Charactersitics  for  Two- Way  Re- 
producing Systems  in  Theaters,  Research  Council,  Academy 
of  Motion  Picture  Arts  &  Sciences 213 

Organization  of  the  Work  of  the  Papers  Committee 

G.  E.  Matthews     217 

Current  Motion  Picture  Literature 225 

1939  Spring  Convention,  Hollywood,  Calif 226 

Society  Announcements 230 


JOURNAL 

OF  THE  SOCIETY  OF 

MOTION  PICTURE  ENGINEERS 


SYLVAN  HARRIS,  EDITOR 

Board  of  Editors 
J.  I.  CRABTREE,  Chairman 

A.  N.  GOLDSMITH  A.  C.  HARDY  H.  G.  KNOX 

J.  G.  FRAYNE  L.  A.  JONES  G.  E.  MATTHEWS 

E.  W.  KELLOGG 


Subscription  to  non-members,  $8.00  per  annum;  to  members,  $5.00  per  annum, 
included  in  their  annual  membership  dues;  single  copies,  $1.00.  A  discount 
on  subscription  or  single  copies  of  15  per  cent  is  allowed  to  accredited  agencies. 
Order  from  the  Society  of  Motion  Picture  Engineers,  Inc.,  20th  and  Northampton 
Sts.,  Easton,  Pa.,  or  Hotel  Pennsylvania,  New  York,  N.  Y. 
Published  monthly  at  Easton,  Pa.,  by  the  Society  of  Motion  Picture  Engineers. 

Publication  Office,  20th  &  Northampton  Sts.,  Easton,  Pa. 
General  and  Editorial  Office,  Hotel  Pennsylvania,  New  York,  N.  Y. 

West-Coast  Office,  Suite  226,  Equitable  Bldg.,  Hollywood,  Calif. 
Entered  as  second  class  matter  January  15,  1930,  at  the  Post  Office  at  Easton, 
Pa.,  under  the  Act  of  March  3,  1879.     Copyrighted,  1939,  by  the  Society  of 
Motion  Picture  Engineers,  Inc. 

Papers  appearing  in  this  Journal  may  be  reprinted,  abstracted,  or  abridged 
provided  credit  is  given  to  the  Journal  of  the  Society  of  Motion  Picture  Engineers 
and  to  the  author,  or  authors,  of  the  papers  in  question.  Exact  reference  as  to 
the  volume,  number,  and  page  of  the  Journal  must  be  given.  The  Society  is  not 
responsible  for  statements  made  by  authors. 


OFFICERS  OF  THE  SOCIETY 

**  President:     E.  A.  WILLIFORD,  30  East  42nd  St.,  New  York,  N.  Y. 
**  Past-President:     S.  K.  WOLF,  RKO  Building,  New  York,  N.  Y. 
**  Executive  Vice-P resident:     N.  Levinson,  Burbank,  Calif. 

*  Engineering  Vice-P  resident:     L.  A.  JONES,  Kodak  Park,  Rochester,  N.  Y. 
**  Editorial  Vice-President:    J.  I.  CRABTREE,  Kodak  Park,  Rochester,  N.  Y. 

*  Financial  Vice-President:     A.  S.  DICKINSON,  28  W.  44th  St.,  New  York,  N.  Y. 
**  Convention  Vice-President:     W.  C.  Kunzmann,  Box  6087,  Cleveland,  Ohio. 

*  Secretary:    J.  FRANK,  JR.,  90  Gold  St.,  New  York,  N.  Y. 

*  Treasurer:     L.  W.  DAVEE,  153  Westervelt  Ave.,  Tenafly,  N.  Y. 

GOVERNORS 
**  M.  C.  BATSEL,  Front  and  Market  Sts.,  Camden,  N.  J. 

*  R.  E.  FARNHAM,  Nela  Park,  Cleveland,  Ohio. 

*  H.  GRIFFIN,  90  Gold  St.,  New  York,  N.  Y. 

*  D.  E.  HYNDMAN,  350  Madison  Ave.,  New  York,  N.  Y. 

*  L.  L.  RYDER,  5451  Marathon  St.,  Hollywood,  Calif. 

*  A.  C.  HARDY,  Massachusetts  Institute  of  Technology,  Cambridge,  Mass. 

*  S.  A.  LUKES,  6427  Sheridan  Rd.,  Chicago,  111. 

**  H.  G.  TASKER,  14065  Valley  Vista  Blvd.,  Van  Nuys,  Calif. 


*  Term  expires  December  31,  1939. 
**  Term  expires  December  31,  1940. 


SOME  TELEVISION  PROBLEMS  FROM  THE  MOTION 
PICTURE  STANDPOINT* 


G.  L.  BEERS,  E.  W.  ENGSTROM,  AND  1.  G.  MALOFF** 


Summary. — -Certain  of  the  characteristics  of  television  have  their  counterparts  in 
motion  pictures,  and  motion  picture  film  and  motion  picture  practice  are  applicable 
to  television.  Some  of  the  problems  and  limitations  pertaining  thereto  are  outlined, 
and  the  following  television-image  characteristics  are  briefly  discussed:  (1)  Number 
of  scanning  lines  and  the  relationship  to  image  size  and  viewing  distance;  (2)  number 
of  frames;  (3)  interlacing. 

The  effect  of  film  and  optical  system  limitations  on  reproduced  television  images  is 
illustrated  by  photographs,  and  curves  are  -given  showing  the  spectral  characteristics 
of  Iconoscopes.  The  screen  color  characteristics  of  Kinescopes  are  also  discussed, 
and  the  overall  range  and  gamma  characteristics  of  a  television  system  are  reviewed. 

The  prime  objective  of  television,  in  common  with  other  pictorial 
arts,  is  to  create  an  illusion.  There  are  certain  limitations  on  how 
good  the  illusion  can  be ;  some  inherent  and  others  dependent  on  the 
state  of  the  art.  Many  of  these  limitations  have  a  counterpart  in 
motion  pictures  and  it  is  the  purpose  of  this  paper  to  review  and  com- 
pare some  of  these  mutual  restrictions. 

PICTURE  DETAIL 

Picture  detail  in  motion  pictures  is  ultimately  determined  by  the 
optical  system  and  the  resolution  of  the  film.  The  factors  determining 
picture  detail  in  television  are  more  complex.  The  frequency  band 
width  limitations  imposed  by  a  single-channel  communication  system 
suitable  for  television  broadcasting  makes  it  necessary  to  divide  the 
scene  arbitrarily  into  elemental  areas  and  transmit  the  information 
representative  of  light  and  shade,  area  by  area  and  line  by  line  until 
the  entire  scene  has  been  scanned.  With  such  a  television  system  the 
basic  factors  determining  picture  detail  are  the  number  of  scanning 
lines,  the  size  of  the  scanning  spot,  the  frequency-band  width,  and 
the  optical  system.  In  practice  the  first  two  factors  are  definitely 

*  Presented  at  the  1938  Fall  Meeting  at  Detroit,  Mich. ;  received  December 
12,  1938. 

**  RCA  Manufacturing  Co.,  Camden,  N.  J. 

121 


122 


BEERS,  ENGSTROM,  AND  MALOFF 


[J.  S.  M.  P.  E. 


related,  since  the  size  of  the  scanning  spot  is  commensurate  with  the 
distance  between  centers  of  scanning  lines.  In  the  television  stand- 
ards of  the  Radio  Manufacturers  Association  scanning  is  expressed  in 
terms  of  the  total  number  of  lines  from  top  to  bottom  from  the  begin- 
ning of  one  frame  to  the  beginning  of  the  next  frame.  Since  in 
a  practical  television  system  both  spot  size  and  frequency-band 
width  are  chosen  on  the  basis  of  the  number  of  scanning  lines,  the 


60  SCANNING  LINES 


120  SCANNING  LINES 


ISO  SCANNING  LINES  240  SCANNING  LINES 

FIG.    1.     Pictures    depicting    characteristics   representative   of    television 
images  for  several  numbers  of  scanning  lines. 

inherent  resolution  of  a  television  system  may  be  expressed  as  the 
number  of  scanning  lines  per  frame. 

Information  has  been  presented  previously  to  indicate  the  degree 
of  entertainment  possible  for  television  images  of  various  numbers  of 
scanning  lines. 1  A  summary  of  this  will  be  presented  here  as  the  first 
step  in  our  analysis  of  how  good  the  television  illusion  will  be.  First, 
we  may  consider  Fig.  1  which  is  made  up  of  four  repetitions  of  the 
same  subject  with  detail  equivalent  to  60,  120,  180,  and  240  scanning 
lines.  From  images  of  this  type  of  motion  picture  film,  from  related 


Feb.,  1939] 


SOME  TELEVISION  PROBLEMS 


123 


tests,  and  from  experience  with  television  systems  Fig.  2  has  been 
made.  This  chart  shows  the  relationship  between  number  of  scanning 
lines  and  picture  size  for  several  viewing  distances.  These  curves 
indicate  what  the  average  eye  demands  for  satisfactory  images  to 
produce  the  illusion  expected  of  television. 

Another  means  for  evaluating  the  resolution  of  a  television  system 
is  to  estimate  its  ability  to  tell  a  desired  story  in  comparison  with 
16-mm.  home  movie  film  and  equipment.  The  result  of  such  a  com- 
parison by  a  number  of  observers  is  that  a  400-  to  500-line  televi- 


500 

/ 

/ 

/ 

8 

.x^ 

/ 

. 

/ 

x 

/ 

UJ 

E 

/ 

/ 

/ 

x 

2 
10    LJ 

/ 

/ 

x 

/ 

X 

^ 

. 

/ 

^ 

/ 

/ 

> 

X 

x 

^ 

x- 

X 

5 

/ 

/ 

^ 

^ 

^ 

/ 

^ 

x- 

7 

^ 

—16  u 

/ 

/ 

/ 

x-^ 

^^ 

2 

> 

y 

/ 

> 

/ 

/ 

X 

X 

/ 

X 

^ 

X 

^^. 

^ 

X- 

x 

^^ 

—  • 

^^-* 

-20 

1- 

1- 

16 

/ 

. 

X 

^ 

/ 

/ 

^ 

^ 

' 

tf 

^ 

^- 

s 

^- 

^—  - 

^-1 
^- 

^^ 

.  —  -~ 

^ 

.  — 

^-- 

-i=- 

100 

PICTURE  HEIGHT    IN    INCHES 

FIG.  2.     Relationship  between  the  number  of  scanning  lines  and 
picture  size  for  several  viewing  distances. 

sion  system  compares  favorably  with  16-mm.  home  movies  in  per- 
mitting observers  to  understand  and  follow  the  action  and  story. 
The  scanning  standard  adopted  by  the  Radio  Manufacturers  Asso- 
ciation is  441  scanning  lines  per  frame. 

FRAME  FREQUENCY  AND  FLICKER 

Television  images  consist  of  rapidly  superimposed  individual  frames 
much  the  same  as  motion  pictures.  In  the  case  of  motion  pictures  a 
group  of  time-related  stills  is  projected  at  a  uniform  rate,  rapid  enough 
to  form  a  continuous  picture  through  persistence  of  vision.  By 
present  methods,  each  frame  of  a  television  image  is  built  up  element 


124  BEERS,  ENGSTROM,  AND  MALOFF          [j.  s.  M.  p.  E. 

by  element  in  some  definite  order  and  these  time-related  frames  are 
reproduced  at  a  rapid  rate. 

In  motion  pictures  the  frame  frequency  determines  how  well  the 
system  will  reproduce  objects  in  motion.  This  has  been  standardized 
at  24  frames  per  second.  In  television  other  factors  than  the  ability 
to  reproduce  motion  have  made  it  necessary  to  use  a  frame  frequency 
of  30  per  second. 

Motion  picture  projectors  commonly  used  are  of  the  intermittent 
type.  The  usual  cycle  of  such  a  projector  is  that  at  the  end  of  each 
projection  period  the  projection  light  is  cut  off  by  a  shutter,  the  film 
is  then  moved  a  step  so  that  the  succeeding  frame  registers  with  the 
picture  aperture,  and  the  shutter  then  opens,  starting  the  next  projec- 
tion period.  This  is  repeated  24  per  second.  Since  projection  at  24 
light-pulses  per  second  with  the  screen  brightness  levels  used  in 
motion  pictures  causes  too  great  a  flicker  effect,  the  light  is  cut  off 
also  at  the  middle  of  the  projection  period  for  each  frame  for  a  time 
equivalent  to  the  period  that  it  is  cut  off  while  the  film  is  moved  from 
one  frame  to  the  next.  This  results  in  projection  at  24  frames  per 
second  with  48  equal  and  equally  spaced  light-pulses.  Such  an  ar- 
rangement provides  satisfactory  results  from  the  flicker  standpoint. 

In  television,  because  of  the  manner  in  which  the  image  is  recon- 
structed, a  continuous  scanning  process,  it  is  not  practicable  to  break 
up  each  light  pulse  further  by  means  of  a  shutter  in  a  manner  similar 
to  that  used  in  the  projection  of  motion  pictures.  We  therefore 
have  in  an  elementary  television  system  a  flicker  frequency  corre- 
sponding with  the  actual  frame  frequency.  This  is  satisfactory  at 
very  low  levels  of  screen  brightness  but  becomes  increasingly  objec- 
tionable as  the  screen  brightness  is  raised. 

In  motion  pictures  the  projector  shutter  opening  in  terms  of  degrees 
for  each  frame  has  an  important  effect  on  the  flicker  characteristics. 
Cathode-ray  tubes — Kinescopes —  are  at  present  the  preferred  means 
for  television  image  reproduction.  In  the  Kinescope  each  element 
of  the  image  on  the  luminescent  screen,  when  excited  by  the  electron 
beam,  fluoresces  and  assumes  a  value  of  brightness  corresponding 
with  the  value  of  the  electron-beam  strength.  Upon  removing  the 
excitation  this  brightness  then  decays  (phosphoresces)  in  an  expo- 
nential manner  and  at  a  rate  dependent  upon  the  screen  material. 
The  phosphorescence  or  persistence  of  the  image  screen  aids  the  per- 
sistence of  vision  of  the  eye  in  viewing  the  reproduced  image.  This 
characteristic  for  one  screen  material  is  shown  in  Fig.  3.  However,  a 


Feb.,  1939] 


SOME  TELEVISION  PROBLEMS 


125 


previously  stated,  far  too  much  flicker  is  present  at  24  or  30  frames 
per  second  for  the  desired  levels  of  screen  brightness. 

A  particular  method  of  scanning  is  therefore  used  to  modify  the 
overall  image  flicker.  This  is  possible  because  scanning  is  a  con- 
tinuous process.  Scanning  may  be  in  equal  horizontal  strips  or  lines 

from  top  to  bottom  in  numerical  order  of  lines  1,  2,  3,  4, 

(progressive  scanning).     This  results  in  one  overall  light-pulse  for 
each  frame.    If  the  procedure  is  modified  so  that  scanning  is  for  the 

first  half  of  one  frame  period  in  the  order  of  lines  1,  3,  5,  7,  9, , 

from  top  to  bottom  of  the  frame  and  for  the  second  half  of  the  frame 
period  in  the  order  of  lines  2,4,6,8,  10, ,  from  top  to  bottom 


Persistence  Characteristic 
of  Kinescope  Screen 


.01    .02     .03 
Time  In  Seconds 


04     .06 


FIG. 


3.     The  phosphorescence  characteristic  of 
a  Kinescope  screen. 


of  the  frame  (interlaced  scanning),  then  the  flicker  effect  of  the  repro- 
duced image  is  changed.  This  method  of  scanning  is  shown  diagram- 
matically  in  Fig.  4.  Each  frame  now  consists  of  two  portions  (two 
fields)  with  respect  to  time:  each  field  composed  of  a  group  of  al- 
ternate lines,  and  the  two  sets  of  alternate  lines  are  properly  staggered 
to  form  a  complete  interlaced  pattern.  In  progressive  scanning  each 
line  flickers  once  per  frame  and  neighboring  lines  differ  in  time  rela- 
tion only  by  the  time  required  for  scanning  one  line.  There  is,  there- 
fore, no  noticeable  inter-line  effect.  In  interlaced  scanning  also  each 
line  also  flickers  once  per  frame,  but  neighboring  lines  differ  in  time 
relation  by  one-half  a  frame  period.  This  results  in  two  flicker  effects, 
an  overall  effect  and  an  inter-line  effect. 


126 


BEERS,  ENGSTROM,  AND  MALOFF 


[J.  S.  M.  P.  E. 


As  previously  stated  a  frame  frequency  of  30  per  second  with  pro- 
gressive scanning  produces  an  intolerable  flicker.  A  frame  frequency 
of  60  per  second  is  certainly  satisfactory  from  the  flicker  standpoint 
but  the  frequency-band  width  required  for  transmission  is  doubled. 
With  interlaced  scanning  at  30  frames,  the  overall  flicker  effect  is 
the  same  as  with  60  frames  progressive  scanning,  and  no  increase  in 
frequency-band  width  is  required.  Each  line  flickers  at  the  rate  of 
30  per  second  and  adjacent  lines  flicker  with  respect  to  each  other, 
since  they  are  scanned  with  a  time-difference  of  Veo  of  a  second.  At 
optimum  viewing  distances  for  television  images  and  for  practicable 
levels  of  screen  brightness  this  inter-line  flicker  is  not  noticeable. 

A  frame-frequency  effect  peculiar  to  television  is  encountered  in  the 
operation  of  cathode-ray  television  receivers  from  an  alternating- 


PROGRESSIVE 
SCANNING 


INTERLACED    SCANNING 


FIG.  4. 


Diagrammatic  illustration  of  progressive  scan- 
ning and  interlaced  scanning. 


current  power-supply  system.  The  effects  of  ripple  voltages  and 
fields  appear  in  the  reproduced  image  in  a  variety  of  forms  and  from 
numerous  sources.  If  the  frame-frequency  differs  from  the  power- 
supply  frequency,  that  is,  differs  except  in  terms  of  integral  multiples 
or  sub -multiples,  then  these  effects  move  across  the  image  at  rates  de- 
pendent upon  the  time-difference  between  the  frame-frequency 
(multiple)  and  the  power-supply  frequency.  This  moving  ripple  pat- 
tern is  almost  as  disturbing  as  flicker  and  the  visual  effects  are  about 
the  same.  Also  for  interlaced  scanning  these  ripple  effects  cause  mov- 
ing displacements  in  the  position  of  alternate  sets  of  lines  and  tend 
to  destroy  the  interlaced  pattern.  If  the  frame-frequency  has  an  in- 
tegral ratio  to  the  power-supply  frequency,  30  frames  for  a  60-cycle 
source,  then  the  effects  are  stationary  on  the  image  and  very  much 
less  pronounced,  thus  making  it  possible  to  obtain  satisfactory  per- 
formance when  using  comparatively  inexpensive  apparatus. 

On  the  basis  of  these  factors  the  Radio  Manufacturers  Association 
has  standardized  interlaced  scanning  with  a  frame-frequency  of  30 
per  second  and  a  field-frequency  of  60  per  second. 


Feb.,  1939]  SOME  TELEVISION  PROBLEMS  127 

Motion  picture  film  is  one  source  of  program  material  for  television. 
With  electronic  scanning  methods  it  is  usual  to  project  an  image  of 
the  film  moving  or  stationary  on  to  some  element  of  the  electronic 
translating  device.  This  may  be  accomplished  by  the  use  of  an  in- 
termittent type  of  projector,  a  continuous  projector  having  an  op- 
tical intermittent,  or  a  system  in  which  the  film  moves  continuously 
with  a  compensating  motion  of  film-image  and  scanning.  The  par- 
ticular method  used  is  partly  determined  by  the  type  of  electronic 
scanning  device.  The  use  of  24-frame  motion  picture  film  to  produce 


SEC. 


SEC. 


SEC: 


SEC. 


DIAGRAM  OF  ONE  COMPLETE  CYCLE  OF  OPERATION    OF  TELE- 
VISION FILM  PROJECTOR 

ENTIRE  CIRCLE  IS  ^  SECOND 

FIG.  5.     Diagram  illustrating  the  3:2  ratio  of  pull-down 
periods   in   a   special   television   film   projector. 

30-frame  television  with  interlaced  scanning  presents  certain  special 
problems. 

In  using  an  Iconoscope  as  the  electronic  translating  device  it  has 
been  customary  to  use  an  intermittent  type  of  projector.  By  utilizing 
the  storage  properties  of  an  Iconoscope,  the  film-image  may  be  pro- 
jected on  to  the  photosensitive  mosaic  during  the  time  between  the 
completion  of  one  field  scanning  and  the  beginning  of  the  next. 
Scanning  then  may  take  place  and  electrical  signals  may  be  obtained 
from  the  mosaic  while  it  is  dark.  Since  the  field  scanning-frequency 
is  at  the  rate  of  60  per  second,  this  means  a  short  period  of  projection 
60  times  a  second  and  of  a  duration  of  approximately  Vsoo  second. 


128  BEERS,  ENGSTROM,  AND  MALOFF  [j.  s.  M.  P.  E. 

With  60  projections  per  second  we  may  hold  one  frame  for  three  pro- 
jection periods — 3/eo  second;  the  next  for  two  projection  periods — 
Veo  second;  the  next  for  three  projection  periods — 3/60  second;  the 
next  for  two  projection  periods.  .  .  .  Thus  by  a  3 : 2  ratio  of  pull- 
down periods  in  an  intermittent  and  by  the  use  of  a  shutter  that  is 
open  only  during  the  vertical  return  time  of  the  scanning,  we  may 
derive  program  material  for  a  30-frame  television  system  from 
standard  24-frame  sound  motion  picture  film,  retaining  the  standard 
film  speed.  This  3 : 2  ratio  of  pull-down  periods  is  illustrated  diagram- 
matically  by  Fig.  5.  Fig.  6  is  a  photograph  showing  two  television 
film  projectors  having  these  characteristics. 

EFFECT  OF  FILM  AND  OPTICAL  SYSTEM  LIMITATIONS  ON  REPRODUCED 
TELEVISION  IMAGES 

In  order  to  verify  previous  conclusions  regarding  the  relative  pic- 
ture detail  capabilities  of  a  441 -line  television  system  and  home  mo- 
tion pictures,  a  test-target  was  photographed  and  reproduced  on 
35-mm.,  16-mm.  and  8-mm.  film.  A  second  purpose  of  making  these 
films  was  to  determine  the  merits  of  the  several  film  sizes  as  sources 
of  television  program  material. 

The  test- target  used  in  making  the  films  consisted  of  twelve  sub- 
stantially identical  major  squares  arranged  in  three  horizontal  rows 
to  form  a  rectangle  having  a  4 : 3  aspect  ratio.  Each  major  square  in- 
cluded four  minor  squares  of  equal  size  but  different  patterns.  Each 
major  square  contained  a  complete  vertical  wedge  of  thirteen  tapered 
bars  (7  black  and  6  white)  which  started  in  the  upper  right-hand  minor 
square  with  a  resolution  calibration  of  100  "resolution  bars"  and  in- 
creased to  200  at  the  bottom  at  that  minor  square.  The  wedge  con- 
tinued in  the  lower  left-hand  minor  square  with  200  at  the  top  and 
300  at  the  bottom.  The  bars  of  the  wedge  were  slightly  curved  so 
that  a  linear  relation  between  distance  along  the  wedge  and  resolu- 
tion could  be  obtained,  facilitating  intermediate  readings.  The  hori- 
zontal wedge  was  similar,  beginning  with  100  at  the  left-hand  side  of 
the  upper-left  minor  square  and  finishing  with  300  in  the  lower-right 
square. 

The  area  surrounding  the  wedge  in  the  two  lower  minor  squares 
of  each  major  square  was  divided  into  four  smaller  areas  which  were 
cross-hatched  to  produce  the  effect  of  different  halftones  from  black 
to  white. 

The  test-films  were  not  intended  to  show  the  maximum  resolution 


Feb.,  1939]  SOME  TELEVISION  PROBLEMS  129 

capabilities  of  the  film  but  to  indicate  the  picture  detail  obtained 
through  present  commercial  methods  for  providing  duplicate  films. 
The  16-mm.  and  8-mm.  films  were  made  from  the  35-mm.  negative 
by  means  of  an  optical  reduction  printer.  Two  16-mm.  and  two  8- 
mm.  prints  were  made.  One  print  of  each  was  made  with  the  custom- 
ary processing  to  give  the  proper  halftone  reproduction.  The  other 
prints  were  made  to  accentuate  the  detail  in  the  wedges  at  some  sacri- 
fice in  halftone  gradation. 

Each  of  the  five  films  was  separately  used  to  produce  a  television 
image  by  projecting  the  test  pattern  on  each  film  on  to  the  mosaic 
of  an  Iconoscope  in  an  experimental  441 -line  television  system.  The 


FIG.  6.     Two  special  television  film  projectors. 

video-frequency  band  employed  by  this  system  was  from  30  to  3,500,- 
000  cycles,  and  the  single  side-band  transmission  this  band-width  is 
well  within  the  channel  limits  that  have  been  tentatively  assigned  for 
television  broadcasting.  For  each  of  the  five  films  a  photograph  was 
taken  of  the  television-image  reproduced  on  the  screen  of  the  Kine- 
scope. These  photographs  are  shown  in  Figs.  7,  8,  9,  10,  and  11.  Fig. 
7  shows  the  result  obtained  from  the  35-mm.  film.  Figs.  8  and  9  illu- 
strate the  images  secured  from  the  16-mm.  film  and  Figs.  10  and  11 
give  the  corresponding  results  with  the  8-mm.  film. 

It  will  be  noted  that  there  is  a  slight  reduction  in  detail  from  the 
image  reproduced  from  the  35-mm.  film  to  that  obtained  from  the 


130 


BEERS,  ENGSTROM,  AND  MALOFF          [j.  s.  M.  p.  E. 


FIG.  7.     Television  image  obtained  from  a  test-chart 
on  35-mm.  film. 


FIG.  8.     Television  image  obtained  from  a  test-chart  on 
16-mm.  film  with  special  processing. 


Feb.,  1939] 


SOME  TELEVISION  PROBLEMS 


131 


FIG.  9. 


Television  image  obtained  from  a  test-chart  on 
16-mm.  film  with  normal  processing. 


FIG.  10. 


Television  image  obtained  from  a  test-chart  on 
8-mm.  film  with  special  processing.   • 


132 


BEERS,  ENGSTROM,  AND  MALOFF 


[J.  S.  M.  P.  E. 


16-mm.  films.    The  loss  in  picture  detail  when  the  -8-mm.  films  are 
used  is,  however,  quite  serious. 

From  these  tests  it  seems  reasonable  to  draw  the  following  con- 
clusions: In  picture-detail  capabilities  a  441 -line  television  system 
compares  favorably  with  16-mm.  home  movies.  If  motion  picture 
film  is  used  to  provide  television  program  material,  satisfactory  re- 
sults may  be  expected  from  35-mm.  film;  a  slight  loss  in  picture-de- 
tail will  result  from  the  use  of  16-mm.  film,  and  the  resolution  ca- 
pabilities of  a  high-definition  television  system  will  not  be  utilized  if 
8-mm.  film  is  used. 


I  •  I —I I— I 

»    «•!    ,«*i  i»aaa 
„  "  PBOB 

i     P— I    4^*1    I 


FIG.  11.    Television  image  obtained  from  a  test-chart  on 
8-mm.  film  with  normal  processing. 


SPECTRAL  CHARACTERISTICS  OF  THE  ICONOSCOPE 

In  motion  picture  work  studio  lighting  and  make-up  technic  are 
dependent  upon  the  color-response  characteristics  of  the  film.  In 
television  the  spectral  response  characteristic  of  the  Iconoscope  con- 
trols these  factors.  This  characteristic  of  an  experimental  Iconoscope 
is  shown  in  Fig.  12.  As  indicated  by  the  curve  this  Iconoscope  gave 
maximum  sensitivity  in  the  blue  end  of  the  spectrum.  The  most 
desirable  Iconoscope  spectral  characteristic  for  a  given  application 
is  dependent  upon  the  light-source  used  to  illuminate  the  scene. 
An  Iconoscope  having  the  characteristic  shown  in  Fig.  12  tends  to 
compensate  for  the  high  red  output  of  the  incandescent  lamps  used 


Feb.,  1939] 


SOME  TELEVISION  PROBLEMS 


133 


in  studio  lighting.  For  outdoor  pick-up  work  a  characteristic  more 
nearly  approximating  that  of  panchromatic  film  is  desired.  The 
spectral  characteristic  of  the  Iconoscope  can  be  varied  to  a  consider- 
able extent  by  the  sensitization  procedure  employed.  Various  spec- 
tral-response characteristics  obtained  in  experimental  Iconoscopes 
have  indicated  that  characteristics  can  be  provided  that  are  com- 
parable to  those  of  panchromatic  and  other  films. 

SCREEN-COLOR  CHARACTERISTICS  OF  KINESCOPES 

The  high-intensity  arc  commonly  used  as  a  light-source  for  motion 
picture  projectors  produces  an  image  that  has  satisfactory  black-and- 
white  characteristics.  In  the  initial  stages  of  cathode-ray  television 


FIG.  12.     Spectral  characteristic  of  an  Iconoscope. 


development,  so  many  serious  limitations  were  present  that  the  green 
color  of  the  image  reproduced  on  a  willemite  screen  was  not  con- 
sidered to  be  particularly  undesirable.  As  television  development 
progressed  and  picture-detail  and  screen  brightness  improved,  the 
green  color  of  the  reproduced  image  became  more  objectionable  and 
development  work  on  luminescent  materials  to  produce  a  black-and- 
white  image  was  started.  Kinescopes  having  luminescent  screens 
giving  black-and-white  pictures  of  adequate  brilliance  are  now  a  com- 
mercial reality.  In  Fig.  13  the  emission  spectra  of  two  screen  mate- 
rials are  shown.  Screens  of  both  materials  will  be  judged  as  white 
if  viewed  separately,  but  when  compared  one  to  the  other  one  will  be 
called  blue-white  and  the  other  ivory-white.  Individual  opinions 


134 


BEERS,  ENGSTROM,  AND  MALOFF 


[J.  S.  M.  P.  E. 


vary  greatly  as  to  which  is  the  best  white  for  television  screens  since 
the  apparent  whiteness  of  a  television  image  is  influenced  by  such 
factors  as  the  image  brightness  and  the  background  lighting  in  the 
room  in  which  the  image  is  viewed.  One  thing  is  certain,  and  that 
is  that  purchasers  of  television  receivers  will  demand  substantially 
black-and-white  images. 

GAMMA   AND  RANGE  IN  TELEVISION 

Those  engaged  in  motion  picture  work  are  quite  familiar  with  two 
terms  that  recently  have  been  given  serious  consideration  in  television, 
i.  e.,  gamma  and  range.  Three  typical  characteristic  curves  of  pic- 
torial reproducing  systems  are  shown  in  Fig.  14.  Curve  (a)  is  for  a 


FIG.  13.     Emission  spectra  of  two  screen  materials. 

contrast  or  gamma  of  unity,  while  curves  (b)  and  (c)  are  for  a  gamma 
of  0.5  and  2.0,  respectively.  The  ranges  available  for  the  image  and 
ranges  of  the  object  that  the  system  can  cover  are  indicated  in  the 
figure.  In  black-and-white  motion  picture  technic,  it  has  become 
standard  practice  to  make  the  overall  object-to-image  contrast  be- 
tween 1.4  and  2  in  order  to  compensate  for  the  lack  of  color.  Tele- 
vision is  also  a  monochromatic  system,  and  it  therefore  seems  de- 
sirable to  follow  the  experience  of  the  motion  picture  industry  and 
produce  television  images  with  a  similar  increase  in  contrast. 

The  "object  brightness  vs.  output  signal"  characteristic  of  the 
Iconoscope  has  been  measured  and  found  to  vary  with  the  amount 
and  distribution  of  light  in  the  object.  However,  it  may  be  stated 
that  in  general  the  Iconoscope  is  a  low-gamma  device — the  value 
varying  between  0.7  and  0.9  for  most  of  the  cases  encountered  in  prac- 


Feb.,  1939] 


SOME  TELEVISION  PROBLEMS 


135 


tice.  The  Kinescope  has  an  inherent  contrast  or  gamma  of  ap- 
proximately 1.5  but  the  saturation  effect  in  the  screen  material  re- 
duces this  to  the  neighborhood  of  1.2.  The  overall  gamma  of  the 
television  system,  so  far  as  the  Kinescope  and  Iconoscope  are  con- 
cerned, is  therefore  substantially  unity.  This  is  quite  satisfactory 
when  motion  picture  film  is  used  to  provide  program  material,  since 
the  contrast  has  already  been  raised  to  a  proper  value  by  an  ex- 
perienced photographer.  An  overall  gamma  of  unity  is  probably  in- 
sufficient when  transmitting  scenes  picked  up  directly  by  the  Icono- 
scope. In  motion  picture  work  gamma  is  controlled  by  the  film 

RANGE   OF    THE   OBJECT  REPRODUCED   BY    THE    SYSTEM 
10,000  :i » 


10  100  1000 

OBJECT  BRIGHTNESS   FT.    LAMBERTS 


10,000 


FIG.  14. 


Three  typical  characteristic  curves  of  pictorial  reproduc- 
ing systems. 


emulsion  and  the  method  and  time  of  development.  In  television  any 
desired  gamma  can  be  obtained  by  varying  the  characteristic  of  one 
of  the  signal  amplifiers  in  either  the  receiver  or  transmitter. 

Although  the  brightness  range  in  television  images  may  be  limited 
in  several  portions  of  the  system  the  present  practicable  limit  is  in  the 
Kinescope.  The  bulb  shape  of  the  Kinescope  is  determined  by  the 
physical  characteristics  necessary  to  withstand  atmospheric  pressure. 
For  this  reason  the  screen  of  a  conventional  Kinescope  has  a  certain 
curvature,  thus  permitting  illuminated  parts  to  throw  light  directly 
on  non-illuminated  parts.  Reflections  may  occur  also  from  other 
portions  of  the  inner  surface  of  the  bulb.  In  addition  to  these  re- 
flections, a  certain  amount  of  light  is  totally  reflected  from  the  glass- 


136  BEERS,  ENGSTROM,  AND  MALOFF  [j.  s.  M.  P.  E. 

air  boundary  and  introduces  a  reduction  of  range  in  details  by  hala- 
tion. These  effects  have  been  reduced  by  blackening  the  inside  walls 
of  the  bulb  and  by  introducing  a  small  amount  of  light-absorbing 
material  in  the  glass  wall.  Conventional  Kinescopes  have  an  avail- 
able range  of  about  50  to  1  for  large  areas  and  10  to  1  in  details.  Ex- 
perimental Kinescopes  have  been  built  in  which  the  luminescent 
screen  is  deposited  on  a  thin  sheet  of  glass  which  is  mounted  inside  a 
transparent  glass  bulb.  Such  tubes  are  capable  of  a  considerably 
greater  range  between  large  areas  and  in  details. 

REFERENCE 

1  ENGSTROM,  E.  W.:  "A  Study  of  Television-Image  Characteristics,"  /.  Soc. 
Mot.  Pict.  Eng.,  XXII  (May,  1934),  p.  290. 

DISCUSSION 

MR.  CRABTREE:  When  televising  an  outdoor  subject,  what  is  the  threshold 
light-intensity  necessary  for  reproduction,  as  compared  with  that  necessary  when 
photographing  with  an  f/2  lens  in  combination  with  the  high-speed  film  emul- 
sion we  now  have  available? 

MR.  BEERS:  With  various  standard  and  special  pick-up  tubes,  we  can  get 
pictures  under  any  lighting  conditions  in  which  you  can  take  pictures  on  film. 

MR.  CRABTREE:  In  other  words,  you  can  reproduce  satisfactorily  a  foot- 
ball game  about  half  an  hour  after  sunset  on  a  rainy  day? 

MR.  BEERS:  Yes.  We  have  obtained  recognizable  pictures  in  which  the 
subject  had  a  surface  brightness  of  1  or  2  foot-candles. 

MR.  CARVER:  I  do  not  understand  whether  you  have  a  flicker  blade  or  not  or 
whether  it  is  or  is  not  necessary. 

MR.  BEERS:  There  is  no  flicker  blade,  as  such.  A  flicker  blade  is  not  neces- 
sary in  television,  due  to  the  way  in  which  we  reconstruct  the  image.  We  pro- 
duce on  the  end  of  the  tube  a  certain  number  of  images  a  second,  and  that  con- 
trols the  flicker.  Nothing  in  the  projector  has  anything  to  do  with  the  manner 
in  which  the  images  are  actually  reproduced. 

MR.  CARVER:  What  do  you  do  when  the  film  is  being  pulled  down?  There 
is  no  picture  on,  and  there  must  be  a  dark  space. 

MR.  BEERS:  You  are  actually  seeing  the  picture  when  the  film  is  being  pulled 
down.  During  the  pull-down  we  scan  the  electrical  image  that  remains  on  the 
mosaic  of  the  Iconoscope.  That  is  when  we  see  the  image  at  the  receiver.  The 
picture  is  projected  when  nothing  is  seen  at  the  receiver.  That  is  the  interval  in 
which  we  transmit  the  synchronizing  signals. 

We  have  two  choices  in  television.  One  is  to  attempt  to  scan  the  picture  on 
the  mosaic  of  the  Iconoscope  during  the  time  the  picture  is  actually  being  pro- 
jected there.  That  means  then  that  we  have  to  pull  down  the  next  frame  of  film 
during  the  time  we  transmit  our  scanning  signals,  which  is  approximately  Vsoo 
of  a  second.  That  imposes  some  physical  requirements  on  a  projector  we  have 
not  been  able  to  meet.  We  have  not  been  able  to  conceive  of  a  projector  on  which 
the  frame  can  be  pulled  into  place  in  1/8oo  second  without  tearing  the  film.  The 


Feb.,  1939]  SOME  TELEVISION  PROBLEMS  137 

easier  way  is  to  pull  the  film  into  place  in  the  gate  during  the  long  interval  of  time 
and  to  project  it  on  the  Iconoscope  mosaic  during  the  short  time  interval;  and 
then  to  scan  it,  while  the  optical  image  is  no  longer  on  the  mosaic,  using  the  elec- 
trical image  that  is  stored  there. 

MR.  FRIEDL:  With  relation  to  standardization,  it  appears  from  these  papers 
that  standards  are  set  on  the  basis  of  electronic  scanning,  determined  by  frame 
frequency,  screen  persistency,  halation,  and  so  on — all  of  which  result  in  the  broad 
frequency-band  which  we  can  admit  is  a  limitation  to  the  general  application  of 
direct  television. 

Reference  was  made  to  mechanical  systems  in  England  and  the  speeds  with 
which  these  systems  operate.  Can  someone  throw  some  light  on  the  question  of 
mechanical  systems  vs.  the  electronic  systems? 

MR.  GOLDSMITH:  The  electronic  systems  offhand  seem  to  be  most  appro- 
priate because  electrons  are  weightless  and  are  readily  controlled  in  flight  and 
form  a  sort  of  "air-brush"  for  painting  pictures  which  can  be  moved  rapidly  and 
readily.  The  only  mechanical  system  that  seems  to  be  seriously  considered 
commercially  today,  at  least  in  England,  is  the  one  Mr.  Kaar  mentioned,  the 
Scophony  system,  in  which  essentially  there  is  a  storage  capacity,  because  of  a 
diffraction  phenomenon  of  standing  waves  of  supersonic  frequency  in  liquid, 
actuated  by  a  vibrating  quartz  crystal.  In  this  system  one  may  scan  a  half  line 
at  a  time  (about  200  picture  elements) . 

This  system,  however,  as  Mr.  Kaar  pointed  out,  employs  a  motor  running  at 
some  30,375  revolutions  per  minute  for  the  high-speed  or  line  deflection  of  the 
spot,  and  a  smaller  motor,  running  at  a  lower  speed,  for  the  frame  scanning,  two 
mirror  systems,  the  supersonic  cell,  some  lenses,  and  a  mercury-vapor  capillary 
lamp  for  the  light-source.  The  picture  produced  is  18  X  24  inches  in  size. 

The  competitive  devices  in  England  are,  for  example,  the  Phillips  receiver 
which  produces  an  18  X  24-inch  picture  by  projection  from  approximately  a  3-inch 
projection  type  cathode-ray  tube. 

The  prices  on  the  British  market  today  are  $850  for  the  Phillips  receiver  pro- 
ducing the  18  X  24-inch  picture,  and  something  over  $1100  for  the  comparable 
Scophony  receiver,  but  nobody  has  given  reliable  data  as  yet  as  to  the  relative 
performance,  life,  and  economics. 

The  receivers  in  England  run  in  the  price  range  from  $125  to  $150  (for  the  pic- 
ture only)  in  a  chair-side  type.  Receivers  for  larger  pictures  run  up  to  $300  or 
$400,  with  top  figures  of  $1200  for  very  large  pictures  (18  X  24  inches)  with 
sound  and  all  sorts  of  extra  attachments,  phonographs,  and  the  like. 

MR.  BEERS:  If  it  were  economically  possible  to  use  24  frames  in  television  it 
would  be  done.  It  is  theoretically  possible,  but  the  increased  amount  of  filtering 
and  shielding  necessary  in  the  receiver  to  make  it  perfectly  satisfactory  from  the 
standpoint  of  eliminating  the  moving  images  resulting  from  the  60-cycle  power- 
supply  system  make  it  economically  impracticable. 

MR.  ROBERTS:  When  scanning  24-frame  motion  pictures  at  60  or  30  cycles, 
do  you  get  any  time-distortion  in  the  presentation  of  the  picture?  Would  there 
be  any  unexpected  effect  as  a  result  of  seeing  one  frame  longer  than  the  one 
before  it? 

MR.  BEERS:  We  have  noticed  no  more  distortion  than  is  normally  noticed  in 
motion  pictures. 


138  BEERS,  ENGSTROM,  AND  MALOFF 

MR.  CRABTREE:  Do  you  think  the  trend  will  be  to  record  directly  by  means 
of  television  scanning,  or  that  the  subjects  will  be  photographed  on  motion  pic- 
ture film  previously  to  scanning?  What  are  the  relative  merits  of  the  two  proc- 
esses— direct  scanning  and  transmission  at  the  scene,  as  against  photographing 
the  scene  and  then  bringing  the  film  to  a  central  transmitting  station? 

MR.  BEERS:  That  is  a  program  problem.  Either  can  be  done.  The  scene 
may  be  taken  on  motion  picture  film  and  then  converted  into  television  program 
material;  or,  as  you  know,  we  have  a  mobile  unit,  which  can  be  taken  out  to  the 
scene  and  there  televise  it  and  transmit  its  picture  directly  by  relay  to  the  trans- 
mitting station. 

MR.  CRABTREE:  When  we  met  in  New  York  we  gained  the  impression  that 
the  actors  hi  the  studios  had  to  work  under  high  temperatures,  with  fans  blowing 
on  them  and  the  make-up  melting  on  their  faces.  In  other  words,  it  seemed  to  be  a 
terrible  ordeal  to  be  a  television  actor.  Would  it  not  be  simpler  to  photograph 
the  scene  and  then  to  transmit  it? 

MR.  GOLDSMITH:     That  condition  has  been  markedly  improved. 

MR.  BEERS:  The  sensitivity  of  the  electronic  pick-up  device  is  being  con- 
stantly improved,  and  as  it  is  improved,  of  course,  this  high-temperature  condi- 
tion that  you  mention  is  reduced.  From  what  little  contact  I  have  had  with 
motion  pictures,  I  am  not  sure  that  conditions  in  that  respect  are  so  vastly  differ- 
ent. The  advantage  in  motion  pictures  is  that  you  can  shoot  a  scene  and  then 
take  time  to  cool  off;  but  in  television,  if  we  wish  to  keep  a  continuous  program 
on  the  air,  the  actors  have  to  stay  under  the  lights  and  suffer.  However,  as  I 
said,  as  the  sensitivity  of  the  pick-up  device  is  improved  the  heat  to  which  the 
actors  are  subjected  will  decrease. 

MR.  SCHLANGER:  Am  I  to  understand  that  the  television  picture  is  equal  in 
quality  to  a  16-mm.  picture,  projected  with  a  250-watt  lamp  and  of  comparable 
size  and  vie  whig  distance? 

MR.  BEERS:  From  the  standpoint  purely  of  picture  detail  they  would  be 
comparable.  I  stated  that  the  film  was  processed  to  produce  duplicate  films. 
If  you  take  standard  home  motion  pictures  on  reversible  film,  you  will  have  some- 
what better  detail,  but  with  ordinary  commercial  processing  of  duplicate  films, 
I  think  that  the  picture  detail  of  the  two  will  be  comparable. 


SOME  PRODUCTION  ASPECTS  OF  BINAURAL  RECORDING 
FOR  SOUND  MOTION  PICTURES* 


W.  H.  OFFENHAUSER,  JR.,  AND  J.  J.  ISRAEL** 

Summary. — The  binaural  effect,  while  of  great  importance  in  our  day-to-day  life, 
has  been  an  unexplored  field  in  our  conventional  monaural  sound  motion  picture 
which,  despite  our  other  technical  advances,  remains  incapable  of  the  quality  called 
sound  ' ' perspective. ' ' 

To  achieve  the  missing  quality,  it  is  important  that  controllably  variable  perspective 
effects  be  attainable  for  the  release  print  from  an  original  having  different  perspective 
quality.  This  is  necessary  not  only  to  make  the  sound  recorded  on  a  three-walled  set 
simulate  the  sound  from  a  six-sided  room,  but  also  to  make  the  recorded  sound  corre- 
spond satisfactorily  with  the  view  portrayed  by  the  camera. 

A  new  system  is  described  that  not  only  fulfills  these  fundamental  requirements,  but 
also  is  capable  of  controllably  producing  at  will  the  various  perspective  and  quality- 
change  effects  in  a  simple  dubbing  operation  from  a  source  having  no  perspective 
characteristics  such  as  a  conventional  monaural  sound-track.  The  new  method  may 
be  introduced  step  by  step  into  our  present  production  system  without  material  obsoles- 
cence of  either  films  or  equipment. 

There  is  a  very  simple  test  that  anyone  can  make  that  should 
conclusively  answer  the  question,  "Why  binaural?"  As  someone 
speaks  to  you,  hold  a  finger  to  one  ear  so  as  to  prevent  sound  from 
entering  that  ear.  You  will  notice  a  marked  difference  in  the  sound; 
it  now  seems  hollow,  and,  if  you  close  your  eyes,  the  voice  that  was 
so  intimately  near  but  a  moment  ago  has  changed  and  now  seems  to 
have  a  different  aspect  which  causes  it  to  appear  indistinct;  and, 
you  are  now  somewhat  confused  as  to  the  true  location  of  the  speaker. 
If,  with  your  eyes  still  closed,  you  remove  the  obstructing  finger  so  as 
to  hear  normally  through  both  ears,  the  voice  of  the  speaker  again 
becomes  distinct  and  takes  on  all  the  characteristics  that  you  regard 
as  natural ;  and  your  speaker  again  seems  human  to  you. 

Despite  the  fact  that  we  are  aware  that  monaural  systems  are 
quite  incapable  of  that  quality  called  perspective,  all  our  commercial 
sound  recording  and  reproducing  systems  in  use  in  motion  pictures 

*  Presented  at  the  1938  Fall  Meeting  at  Detroit,  Mich. ;  received  October  3, 
1938. 

**  New  York,  N.  Y. 

139 


140  W.  H.  OFFENHAUSER,  JR.,  AND  J.  J.  ISRAEL  [j.  s.  M.  p.  E. 

today  are  monaural  or  one-eared  systems.  Technical  development 
of  monaural  systems  has  been  pushed  ahead  by  the  industry  at  full 
speed;  the  success  of  these  efforts  has  been  reflected  in  renewed 
energy  in  the  development  of  the  newest  advance  which  no  doubt 
will  soon  take  its  place  commercially  in  the  industry — binaural 
sound  recording  and  reproduction.  Binaural  recording,  like  stere- 
oscopic pictures,  has  a  long  history  of  achievement  as  well  as  dis- 
illusionment. The  difficulty  encountered  in  approaching  both  sub- 
jects has  the  same  character;  it  is  necessary  for  us  to  understand 
human  reactions  before  we  can  attempt  even  to  consider  technical 
solutions.  This  analysis  of  the  human  phase  at  first  glance  seems 
simple ;  yet,  closer  inspection  reveals  that  the  more  we  look  into  the 
subject,  the  less  we  appear  to  comprehend. 

Before  any  methods  or  apparatus  can  be  considered,  it  is  necessary 
to  approach  the  problem  in  the  manner  in  which  we  approach  any 
problem  of  evaluation;  we  must  first  determine  our  hypotheses  and 
our  axioms;  and  only  after  these  are  fully  decided  upon  may  we 
approach  our  theorems.  This  must  be  done  with  fear  and  trem- 
bling ;  every  step  in  the  formulation  of  the  hypotheses  and  axioms 
must  necessarily  involve  compromise  as  it  will  be  found  that  no  two 
workers  in  the  field  can  agree  completely  upon  either  the  premises  or 
the  objectives. 

We  must  leave  to  the  psychologists  the  problem  of  evaluation  of 
human  reactions.  In  every  step  of  our  reasoning  we  must  keep  in 
mind  the  fact  that  a  motion  picture  audience  is  a  group  of  customers 
who  take  nothing  tangible  home  with  them  despite  the  fact  that  they 
pay  something  tangible  for  what  they  receive.  It  is  the  purpose  of 
the  motion  picture  to  create  an  illusion  by  whatever  means,  and  the 
most  successful  motion  picture  is  the  one  that  creates  the  best  illusion. 

Our  psychologists  have  pointed  out  for  centuries  that  reality  and 
our  concepts  of  it  may  have  little  to  do  with  one  another.  We  have 
been  told  repeatedly  that  only  the  abstract  can  be  perfect :  the  more 
real  (in  the  sense  of  physical  reality)  a  thing  becomes,  the  more  notice- 
able are  its  flaws.  The  same  may  be  said  to  be  true  of  motion  pic- 
tures. 

This  point  has  been  amply  demonstrated  time  and  again.  In  the 
recording  of  sound,  for  example,  we  would  not  always  care  to  have 
the  noise  of  a  collapsing  building  reproduced  for  us  in  our  theaters  at 
its  original  volume,  particularly  if,  after  the  building  collapsed  we 
wished  to  hear  dialog  between  two  of  our  principal  characters.  The 


Feb.,  1939]  BlNAURAL  RECORDING  141 

perfect  picture  would  be  one  that  would  create  the  illusion  of  the 
falling  building  without  causing  the  physical  discomfort  that  would 
result  if  we  were  to  hear  the  noise  in  all  its  original  intensity. 

In  motion  pictures  it  is  the  illusion  produced  by  the  facile  and 
artistic  combination  of  both  picture  and  sound  that  is  the  desideratum 
of  a  "good  show."  This  is  forcefully  pointed  out  when  we  consider  a 
hypothetical  case  that  could  well  take  place.  Suppose  one  were  to 
make  a  sound  record,  without  picture,  of  the  waves  breaking  on  the 
beach  at  Atlantic  City.  For  the  sake  of  illustration  let  us  boldly 
assume  that  we  are  able  both  to  record  and  reproduce  this  sound 
without  any  loss  of  fidelity  whatever.  We  will  project  this  sound 
under  ideal  conditions  for  a  group  of  Iowa  Agricultural  College 
students,  most  of  whom  have  never  been  to  the  seashore.  What  will 
be  the  reactions  of  this  audience  after  a  long  afternoon  spent  in  the 
hot  sun  studying  Japanese  beetle  control  ?  There  is  only  one  answer : 
our  recording  accomplished  with  such  technical  perfection  is  nothing 
more  than  a  meaningless  noise  that  is  at  first  quite  boring,  and,  after 
a  few  seconds,  a  plain  nuisance  from  which  our  students  will  seek 
escape. 

Binaural  recording  is  something  that  we  ought  to  have  in  motion 
pictures  today.  It  will  immeasurably  enhance  the  illusion  when 
properly  applied,  as  has  been  indicated  by  the  remarkable  sound 
illusions  produced  by  the  engineers  of  the  ERPI  and  allied  groups. 
A  motion  picture  director  with  an  imaginative  mind  would  revel  in  the 
delightful  possibilities  of  creating  illusions  with  even  today's  com- 
mercial motion  picture  and  today's  demonstrated  results  in  binaural 
sound  transmission.  With  stereoscopic  color  pictures  and  binaural 
recording,  our  director  could  well  indulge  in  an  artistic  orgy. 

Motion  pictures  with  binaural  sound  can  hardly  be  considered 
visionary  today:  many  years  before  sound  was  introduced  com- 
mercially into  our  motion  picture  theaters,  patents  had  been  granted 
to  far-seeing  inventors  on  the  combination  of  motion  pictures  with 
binaural  sound.  The  binaural  patent  that  is  the  forerunner  of  them 
all  is  that  of  Rosenberg,  applied  for  in  Great  Britain  on  October  25, 
1911  (Brit.  Pat.  No.  23,620).  Rosenberg's  appreciation  of  the 
problem  is  truly  remarkable  and  his  clarity  can  well  be  considered  a 
model  for  others  to  follow. 

It  is  doubtful  that  Rosenberg's  reduction  to  practice  consisted  in 
much  more  than  the  filing  of  his  patent  application.  Surely  there 
was  little  that  he  could  do  at  the  time  in  making  models  of  his  in- 


142  W.  H.  OFFENHAUSER,  JR.,  AND  J.  J.  ISRAEL  [J.  S.  M.  P.  E. 

vention.  The  vacuum  tube  was  practically  unknown  outside  of  a 
very  few  research  laboratories.  The  quality  possible  with  existing 
microphones  and  loud  speakers  was  such  as  to  discourage  any  inven- 
tor with  far-reaching  ideas.  It  is  indeed  a  tribute  to  the  man's  con- 
fidence in  his  ideas  that  he  prosecuted  his  patent  to  a  successful  con- 
clusion, and  a  tribute  to  the  British  Patent  Office  that  his  patent  was 
passed  to  issue. 

If  one  reviews  the  advance  of  the  art  since  Rosenberg,  it  does  seem 
fair  to  say  that  most  inventions  since  then  have  been  primarily 
adaptations  of  currently  existing  mechanisms  to  the  art  as  disclosed 
by  him.  There  has  been  little  of  theoretical  value  added  unless  it  be 
the  concepts  of  Kuechenmeister  dealing  with  the  delay  effect.  (Brit. 
Pat.  No.  238,372;  applied  for  Aug.  12,  1924.)  This  currently  popu- 
lar concept  as  applied  to  binaural  recording  on  film  is  shown  in 
Kuechenmeister  British  patent  258,864,  which  had  a  convention  date 
of  Sept.  22,  1925.  Certain  other  inventions  may  be  classified  as 
frequency  characteristic  variation;  still  others  have  shown  multi- 
plicities of  microphones,  multiplicities  of  amplifiers,  multiplicities  of 
recording  means,  and  multiplicities  of  loud  speakers ;  others  disclose 
dummy  heads  and  their  equivalents  with  microphonic  ears.  All 
these  things  have  been  combined  with  motion  pictures,  television, 
radio,  phonographs,  broadcasting,  as  well  as  an  almost  infinite  variety 
of  other  things  worthy  of  passing  mention.  In  the  motion  picture 
field,  these  have  even  been  combined  with  "wandering"  sound- 
tracks and  other  inventions  of  similar  nature. 

The  improvement  that  has  occurred  in  the  art  of  motion  picture 
production  since  sound  was  first  introduced  commercially,  can  be 
summarized  briefly  as  an  improvement  in  technic.  In  the  physical 
stages  of  production,  there  exist  both  shooting  and  editing,  and  it  can 
truly  be  said  that  today  a  picture  is  made  in  the  cutting  room.  The 
editing  aspect  is  almost  daily  acquiring  greater  and  greater  im- 
portance. 

The  addition  of  anything  new  to  sound  motion  picture  production 
must  consider  the  adaptability  of  the  new  element  to  the  editing 
process  as  well  as  to  the  shooting  process.  If  the  new  added  element 
involves  little  or  no  added  complexity  in  the  shooting  process,  so 
much  the  better.  Stage  time  costs  money  as  all  producers  know 
only  too  well.  If  the  new  added  element  further  involves  but  a  small 
added  complexity  in  the  editing  process,  it  can  be  introduced  readily 
provided  the  scheduled  editing  time  and  expense  are  not  unduly 


Feb.,  1939]  BlNAURAL  RECORDING  143 

increased.  This  is  important  since  the  present  tendency  is  to  limit 
stage  operations  and  to  delegate  more  and  more  work  to  the  editing 
process.  H.  G.  Tasker's  description  of  the  production  process  given 
to  the  Society  at  the  last  Hollywood  meeting  points  out  indirectly  the 
importance  of  editing  when  the  procedure  he  described  is  compared 
with  what  we  now  consider  the  antiquated  methods  of  1928  and  1929 
when  pictures  were,  for  the  most  part,  made  on  the  shooting  stage. 

At  first  glance,  the  application  of  binaural  recording  to  motion 
pictures  would  seem  to  hold  unknown  terrors  for  the  production 
supervisor.  It  would  seem  that  there  would  be  no  assurance  that  the 
desired  effect  could  be  produced  at  all  even  if  it  were  accurately  de- 
fined; and,  for  that  reason,  shooting  delays  would  seem  almost 
certain  to  occur  due  to  the  introduction  of  the  new  technic,  result- 
ing in  prohibitive  shooting  costs. 

Let  us  consider  such  a  simple  scene  as  that  of  a  talking  actor  passing 
through  a  doorway.  Our  actor  performs  in  a  three-walled  set  located 
on  a  shooting  stage.  It  is  difficult  to  record  scientifically  "perfect" 
sound  for  such  a  sequence.  In  the  first  place,  the  sound  heard  must 
correspond  with  the  picture  seen;  the  picture  is  dependent  upon 
camera  location  and  lens  size.  In  the  second  place,  even  assuming 
that  we  could  make  the  sound  and  picture  correspond,  it  would  still 
be  impossible  to  produce  by  usual  straightforward  methods  a  sound 
record  having  the  characteristics  of  a  six-sided  room  when  the  record- 
ing is  actually  made  in  a  three- wall  set.  A  director  would  "throw 
up  his  hands"  as  soon  as  the  actor  started  through  the  doorway; 
the  extremely  rapid  swinging  of  the  microphones  suspended  on  the 
boom  necessary  to  produce  a  suitable  acoustic  spatial  effect  would  be 
entirely  impracticable  in  a  set  of  present-day  construction.  Under 
the  circumstances,  it  is  rather  impracticable  to  attempt  scientifically 
"perfect"  recordings  especially  when  it  is  an  illusion  that  we  are 
trying  to  produce,  and  we  are  already  aware  that  reality  and  our  con- 
cept of  it  may  differ  radically. 

The  tricks  of  the  sound-effects  men  are  a  case  in  point.  It  is  indeed 
a  revelation  to  the  average  movie-goer  to  view  the  various  gadgets 
used  by  these  artists  in  the  production  of  their  convincing  sound 
illlusions.  In  spite  of  our  innate  desire  for  scientific  truth,  we  do  not 
exhibit  to  the  public  gaze  the  props  that  form  such  an  important 
part  of  our  artistic  legerdemain;  our  attitude,  rather,  is  to  produce 
the  best  illusion  we  know  how  and  to  ignore  entirely  the  means  of  its 
production.  The  mechanical  contrivances  backstage  would  not  add 


144  W.  H.  OFFENHAUSER,  JR.,  AND  J.  J.  ISRAEL  [j.  s.  M.  P.  E. 

to  the  illusion  if  they  were  exposed  to  public  view;  we  go  even  so  far 
as  to  conceal  them  willfully. 

Why,  then,  in  the  case  of  an  actor  going  through  a  doorway,  is  it 
necessary  for  us  to  swing  microphones  at  a  prodigious  rate  merely  to 
create  a  scientifically  "perfect"  recording,  when  it  is  possible  not 
only  to  produce  the  desired  illusion  by  far  simpler  means  but  also  to 
do  it  without  any  especially  serious  change  in  shooting  technic? 
This  is  all  the  more  forcefully  pointed  out  when  it  is  realized  that  our 
scientifically  "perfect"  recording  can  at  best  produce  only  the  wrong 
illusion,  because  we  are  recording  in  a  three-walled  set. 

Suggestions  have  appeared  in  the  prior  art1-2'3-4  from  time  to  time 
that  a  procedure  is  possible  for  moving  sound  around  in  reproduction 
without  any  movement  of  the  sound  on  the  shooting  stage.  A 
typical  instance  is  Jones  in  U.  S.  Pat.  No.  1,855,146  and  the  various 
divisions  of  the  invention  there  disclosed.  In  the  vernacular,  Jones 
may  be  said  to  disclose  "the  wriggling  of  microphonic  ears  in  a  dummy 
head."  Regardless  of  the  means,  there  has  been  little  serious  con- 
sideration of  this  feature  of  moving  sound  around  artificially  in  test 
and  other  stereophonic  and  binaural  films  that  have  been  exhibited  as 
samples  of  the  binaural  art.  If  the  binaural  art  is  to  be  fitted  into 
everyday  sound  motion  picture  production,  it  must  consider  for  the 
release  print  the  production  of  acoustic  effects  that  were  not  produced 
on  the  shooting  stage. 

Rosenberg  clearly  described  the  requirements:" — the  necessity  for 
communicating  to  the  spectator  an  impression  of  constant  coincidence 
between  the  actual  and  the  visually  apparent  sources  of  a  train  of 
sounds  as  reproduced  is  due  to  the  fact  that  .  .  .  the  perfection  of  the 
illusion  will  be  impaired  unless  those  sounds  which  represent  the 
voice  of  the  performer  are  at  each  instant  made  to  appear  as  if  they 
actually  proceeded  from  whatever  spot  the  speaker  is  himself  visually 
represented  as  momentarily  occupying."5 

Jones6  has  suggested  that  the  apparent  location  o  a  sound  in  space 
can  be  shifted  about  by  the  "wriggling"  of  the  microphonic  ears  in 
the  dummy  head  that  he  disclosed.  His  theory  is  based  on  the  delay 
concept  that  Kuechenmeister7  has  aptly  described — "means  intended 
to  convert  into  sound  the  record  optically  produced  on  the  talking 
film  are  so  arranged  with  respect  to  the  said  record  on  the  film  that 
the  sound  reproducer  or  reproducers  connected  to  the  said  means  will 
repeat  the  sound  at  a  small  interval  of  time."  Kuechenmeister  did 
describe  this  delay  effect  as  a  phase-difference  and  expressed  his 


Feb.,  1939]  BlNAURAL  RECORDING  145 

phase  difference  as  "being  about  Vso  to  Vs  of  a  second."  Jones 
describes  his  delay  effect  as  a  phase-difference  also,  only  he  measures 
his  delay  in  what  he  calls  "sound-inches";  a  coined  term  to  indicate 
the  time-interval  required  for  sound  to  travel  a  distance  of  one  inch. 

Investigators  in  the  field  are  not  agreed  as  to  whether  phase- 
difference  has  something  to  do  with  the  problem  or  not;  the  point  at 
issue  seems  to  be  the  definition  of  phase-difference  as  applied  to  this 
particular  problem.  A  number  of  investigators  are  quite  definite  and 
state  with  apparent  assurance  that  phase-difference  has  nothing 
whatever  to  do  with  the  case  and  that  it  is  amplitude-difference  that 
is  the  crux  of  the  question.  Other  investigators  state  with  apparently 
equal  assurance  that  phase-difference  is  the  crux  of  the  question, 
despite  the  fact  that  much  of  the  prior  art  has  accepted  the  ampli- 
tude-difference theory.  Still  others  straddle  the  fence  and  propose 
the  conversion  of  phase-differences  into  amplitude-differences, 
particularly  in  the  low-frequency  portion  of  the  audio  spectrum. 
To  an  impartial  student,  such  a  condition  is  indicative  of  lack  of 
agreement  upon  hypotheses.  As  Mark  Twain  put  it,  "An  argument 
arises  when  people  use  the  same  words  to  describe  different  things  or 
when  they  use  different  words  to  describe  the  same  thing." 

It  looks  as  if  we  engineers  are  stepping  out  of  our  field  a  bit  at  this 
stage  in  attempting  to  define  phase-difference.  We  are  not  specific  as 
to  whether  our  aural  impressions  are  air-borne  through  a  common 
medium  or  whether  the  binaural  impressions  are  separately  conveyed. 
In  the  motion  picture  business  we  are  of  course  interested  in  the 
former;  the  research  man  has  primary  interest  in  the  latter  as  a  step 
in  the  fuller  understanding  of  the  former.  It  is  not  our  privilege  to 
decide  how  the  two  impressions  received  at  the  two  ears  are  integrated 
in  the  brain.  That  is  the  province  of  the  physiologist  and  the  psychol- 
ogist and  we  are  getting  ourselves  into  a  limitless  morass  unless  we 
approach  this  problem  in  the  same  scientific  manner  in  which  we 
approach  our  other  technical  problems. 

Our  purpose  is  to  create  an  illusion  and  the  most  successful  motion 
picture  is  the  one  that  creates  the  best  illusion.  Let  us  then  put  aside 
for  the  time  being,  at  least,  the  question  of  phase-difference  vs.  ampli- 
tude-difference and  do  a  bit  of  listening  with  our  ears.  When  we 
have  once  decided  what  creates  the  illusion  we  want,  let  us  then  again 
attempt  to  analyze  the  means  of  its  production,  after  the  psychologists 
have  told  us  what  we  are  now  in  dire  need  of  knowing. 

There  are,  of  course,  a  number  of  points  of  agreement.     We  can 


146  W.  H.  OFFENHAUSER,  JR.,  AND  J.  J.  ISRAEL  [J.  S.  M.  P.  E. 

agree  that  two-eared  hearing  as  we  experience  it  in  everyday  life  is 
better  than  one-eared  hearing.  We  can  also  agree  that  in  one-eared 
hearing,  reverberation  plays  an  important  part  in  conveying  the 
depth  impression.  We  can  also  agree  that  if  we  have  one  sound 
record  and  reproduce  it  as  Kuechenmeister  did  through  two  chan- 
nels, controlling  the  delay  of  one  cliannel  with  respect  to  the  other, 
certain  effects  are  produced  that  to  many  persons  are  pleasing.  We 
can  also  agree  that  the  practical  demonstrations  of  auditory  per- 
spective given  by  the  ERPI  engineers  produced  sound  results  that 
were  as  desirable  as  they  were  startling.  All  that  remains  is  to 
apply  these  effects  to  the  production  of  sound  motion  pictures  in  a 
practicable  and  economic  manner. 

If  we  ignore  for  the  moment  reverberation  effects,  it  can  be  fairly 
said  that  it  is  our  present  practice— and  a  very  desirable  one — to 
record  all  our  sound  with  relatively  close  microphone  placement.  It 
is  in  this  manner  that  we  are  able  to  reduce  extraneous  noise  to  a 
minimum.  If,  then,  we  can  continue  so  to  record  and  later  produce 
in  the  dubbing  room  the  desired  perspective  and  spatial  effects  with- 
out any  increase  in  noise-level  other  than  that  due  to  the  addition  of 
the  usual  re-recording  step  in  the  production  of  the  final  print,  we 
then  have  a  system  which  from  an  engineering  viewpoint  is  ideal,  in 
that  it  allows  the  highest  signal-to-noise  ratio  and  also  the  attainment 
of  the  desired  dramatic  perspective  and  spatial  effects  in  the  dubbing 
room,  where  the  cost  of  retakes  is  so  small  relative  to  those  of  the 
shooting  stage  that  it  is  possible  to  avoid  compromise  with  quality  in 
the  sound  portion  of  the  presentation. 

With  the  system  to  be  demonstrated,  it  is  possible  not  only  sub- 
stantially to  achieve  these  desiderata  but  also  to  produce  quite 
similar  effects  using  only  a  standard  single  channel  in  recording,  and 
nevertheless  effecting  the  perspective  control  in  the  dubbing  room 
much  in  the  manner  of  a  binaural  recording.  As  is  to  be  expected, 
the  quasi-binaural  system,  as  we  may  call  it,  is  not  quite  as  con- 
vincing as  the  full  binaural  system.  When  the  illusion  is  aided  by  the 
picture,  however,  it  is  doubtful  whether  the  average  audience  would 
notice  the  difference,  particularly  if  the  quasi-binaural  films  were 
released  in  the  transition  period  from  monaural  to  binaural. 

The  change  from  monaural  to  binaural  films  in  reproduction  can  be 
made  without  difficulty.  In  projecting  binaural  films,  we  can  select 
either  one  track  or  the  other  or  possibly  mix  the  two  by  scanning 
both  with  the  same  light-beam  in  a  conventional  monaural  projector. 


Feb.,  1939]  BlNAURAL  RECORDING  147 

Thus  a  film  printed  for  binaural  release  can  be  projected  satisfactorily 
on  a  conventional  projector  and  at  the  same  time  be  suited  to  pro- 
jection on  a  binaural  projector.  There  is  no  technical  reason  why  all 
films  should  not  be  printed  for  binaural  release. 

In  the  production  studio,  the  change  to  binaural  can  be  made  with 
equal  facility.  Since  it  is  possible  to  dub  a  monaural  recording  for 
binaural  release,  films  on  the  shelf  are  not  made  obsolete  merely 
because  a  new  improvement  in  sound  recording  has  arrived. 

The  stage-shooting  technic  for  binaural  recording  requires  no  serious 
change  from  the  present  technic  since  the  perspective  control  portion 
of  the  process  is  to  be  effected  primarily  in  the  dubbing  room.  On 
the  sound-stage,  the  sound  may  be  recorded  much  as  it  is  today  with 
the  possible  modification  that  somewhat  less  microphone  manipula- 
tion may  prove  desirable.  As  the  experience  gained  in  production 
increases,  the  advantageous  deviations  from  the  present  procedure 
may  be  studied  and  later  introduced  as  their  peculiarities  become 
better  understood.  The  change  to  the  new  system  requires  no  drastic 
revisions  of  either  operating  technic  or  apparatus  other  than  the 
eventual  introduction  of  the  necessary  second  channel. 

There  seems  to  be  but  little  agreement  upon  the  acoustic  effects 
desired.  If  we  consider  the  requirements  as  defined  by  Rosenberg 
we  can,  in  all  likelihood,  agree  that  left-right  movement  is  desirable ; 
that  some  convincing  impression  of  the  distance  of  a  sound-source 
from  the  scene  represented  is  likewise  desirable;  and  that  possibly 
some  impression  of  the  directness  with  which  the  sound  is  presumed 
to  approach  the  listener  is  desirable.  On  the  other  side  of  the  Atlan- 
tic, this  last  is  known  as  the  "round- the-corner"  effect.  As  in  the 
past,  our  procedures  will,  no  doubt,  best  be  worked  out  by  rule  of 
thumb. 

The  system  to  be  demonstrated  is  not  limited  to  motion  pictures; 
it  may  be  applied  in  any  field  where  it  is  desired  to  transmit  im- 
pressions by  means  of  sound.  Some  of  the  effects  that  have  been 
produced  with  the  system  include  variation  of  the  apparent  recording 
room  size  from  very  small,  say,  1000  cubic  feet,  to  very  large,  say, 
500,000  cubic  feet.  Also  important  is  the  simultaneous  yet  inde- 
pendent movement  in  reproduction  of  one  sound-source  with  respect 
to  another.  Essentially  independent  left-right  movement  as  well  as 
close-up  and  distant  perspective  has  also  been  attained.  All  these 
effects  were  produced  in  reproduction  without  movement  of  the 
actual  sound-sources  with  respect  to  the  microphones.  In  motion 


148  W.  H.  OFFENHAUSER,  JR.,  AND  J.  J.  ISRAEL  [j.  s.  M.  p.  E. 

pictures,  the  application  of  such  a  system  can  appreciably  increase 
editing  latitude  and  improve  dramatic  effectiveness. 

The  authors  then  proceeded  to  demonstrate  the  system.  A  conventional  single- 
channel  disk  reproducing  turntable  was  connected  to  the  perspective-control  device  at 
the  rear  of  the  room.  Two  channels  from  this  device  fed  independently  to  two  loud 
speakers  placed  side  by  side  on  a  platform  directly  before  the  audience.  No  movement 
of  the  speaker  with  respect  to  the  microphone  or  the  surroundings  occurred  when  the 
record  was  made;  the  perspective  or  binaural  effect  was  produced  entirely  by  the 
manipulation  of  the  prespective  control  in  the  rear  of  the  room.  The  sounds  appeared 
to  move  from  one  side  of  the  room  to  the  other,  upward  and  downward,  or  backward  and 
forward,  as  the  controls  were  manipulated. 

REFERENCES 

1  BEERS,  G.  L.:     U.  S.  Pat.  2,098,561  (Nov.  9,  1937). 

2  HALSTEAD,  W.  S. :     U.  S.  Pat.  2,022,665  (Dec.  3, 1935). 
HAMMOND,  J.  H.,  JR.  :     U.  S.  Pat.  2,008,712  (July  23,  1935). 
HAMMOND,  J.  H.,  JR.  :     U.  S.  Pat,  2,060,204  (Nov.  10,  1936). 

3  JONES,  W.  B. :     U.  S.  Pat.  1,855,149  (Apr.  19,  1932). 
DOOLITTLE,  F.  M. :     U.  S.  1,817,177  (Aug.  4,  1931). 

4  PATTERSON,  W.  M.:     U.  S.  Pat.  1,994,920  (Mar.  19,  1935). 
DOUDEN,  W.  L.:     U.  S.  Pat.  2,124,030  (July  19,  1938). 

6  ROSENBERG,  A.:     Brit.  Pat.  23,620  (Oct.  25,  1911)  (application  date). 

6  JONES,  W.  B. :     U.  S.  Pat.  1,855,149,  p.  3,  line  36  et  seq. 

7  KUECHENMEISTER,  J. :     Brit.  Pat.  (Aug.  12,  1924)  (application  date). 

(Patent  dates  listed  are  dates  of  issue  unless  otherwise  specified.} 

BIBLIOGRAPHY 

STEWART,  G.  W.:  "The  Intensity  Factors  in  the  Binaural  Localization  of 
Sound,"  Phys.  Rev.,  Series  1,  34  (1912),  p.  76  (abstract). 

"The  Significance  of  Intensity-Sum  in  Binaural  Localization,"  Phys.  Rev., 
Series  2,  II  (1913),  No.  l,p.  72. 

"The  Character  of  Interaural  Sound  Conduction  by  Binaural  Beats,"  Phys. 
Rev.,  Series  2,  III  (1914),  No.  2,  p.  146. 

"Phase  Relations  in  the  Acoustic  Shadow  of  a  Rigid  Sphere;  Phase  Difference 
at  the  Ears,"  Phys.  Rev.,  Series  2,  IV  (1914),  No.  3,  p.  252. 

"The  Intensity  Factor  in  Binaural  Localization  and  an  Extension  of  Weber's 
Law,"  Phys.  Rev.,  Series  2,  IX  (1916),  No.  4,  p.  338  (abstract). 

"The  Functions  of  Intensity  and  Phase  in  the  Binaural  Location  of  Pure 
Tones.  An  Experimental  and  Theoretical  Investigation  of  Binaural  Beats," 
Phys.  Rev.,  Series  2,  IX  (1917),  No.  6,  p.  338  (abstract). 

"Binaural  Beats,"  Phys.  Rev.,  Series  2,  IX  (1917),  No.  6,  p.  502. 

"The  Secondary  Intensity  Maxima  in  Binaural  Beats,"  Phys.  Rev.,  Series  2, 
IX  (1917),  No.  6,  p.  509. 

"The  Theory  of  Binaural  Beats,"  Phys.  Rev.,  Series  2,  IX  (1917),  No.  6,  p. 
514. 


Feb.,  1939]  BlNAURAL  RECORDING  149 

"The  Functions  of  Intensity  and  Phase  in  the  Binaural  Location  of  Pure 
Tones,"  Phys.  Rev.,  Series  2,  XV  (1920),  No.  3,  p.  248  (abstract). 

"The  Function  of  Intensity  and  Phase  in  the  Binaural  Location  of  Pure  Tones," 
Phys.  Rev.,  Series  2,  XV  (1920),  No.  6,  pp.  425,  432. 

SIMPSON,  M.:  "Experiments  in  Binaural  Phase  Difference  Effect  with  Pure 
Tones,"  Phys.  Rev.,  Series  2,  XV  (1920),  No.  6,  p.  421. 

HARTLEY,  R.  V.  L. :  "The  Function  of  Phase  Difference  in  the  Binaural  Loca- 
tion of  Pure  Tones,"  Phys.  Rev.,  Series  2,  XIII  (1919),  No.  6,  p.  373. 

MAXFIELD,  J.  P.:  "Some  Physical  Factors  Affecting  the  Illusion  in  Sound 
Motion  Pictures,"  /.  Acoust.  Soc.  Amer.,  Ill  (July,  1931),  No.  1,  p.  69. 

FLETCHER,  H.:  "Auditory  Perspective — Basic  Requirements,"  Electrical 
Engineering,  53  (Jan.,  1934),  No.  1,  p.  9. 

STEINBERG,  J.  C.,  AND  SNOW,  W.  B.:  "Auditory  Perspective — Physical  Fac- 
tors," Electrical  Engineering,  53  (Jan.,  1934),  No.  1,  p.  12. 

MAXFIELD,  J.  P.:  "Demonstration  of  Stereophonic  Recording  with  Motion 
Pictures,"  /.  Soc.  Mot.  Pict.  Eng.,  XXX  (Feb.,  1938),  No.  2,  p.  131. 

"Pick-Up  for  Sound  Motion  Pictures  (including  Stereophonic),"  /.  Soc.  Mot. 
Pict.  Eng.,  XXX  (June,  1938),  No.  6,  p.  666. 

HARTLEY,  R.  V.  L.,  AND  FRY,  T.  C.:  "The  Binaural  Location  of  Complex 
Sounds,"  Bell  Syst.  Tech.  J.  (Nov.,  1922),  p.  33. 

MAXFIELD,  J.  P. :  "Some  Physical  Factors  Affecting  the  Illusion  in  Sound  Mo- 
tion Pictures,"  /.  Acoust.  Soc.  (July,  1931),  p.  69. 

"Auditory  Perspective"  (a  symposium),  Elect.  Eng.  (Jan.,  1934),  pp.  9,  216; 
Bell  Syst.  Tech.  J.  (April,  1934),  p.  239. 

CHAPIN,  E.  K.,  AND  FIRESTONE,  F.  A.:  "  The  Influence  of  Phase  on  Tone 
Quality,"  /.  Acoust.  Soc.  Amer.,  V  (Jan.,  1934),  p.  40. 

DISCUSSION 

MR.  OFFENHAUSER:  There  is  one  feature  in  particular  to  which  we  would  like 
to  call  your  attention  and  that  is  the  fact  that  it  is  now  possible  to  make  the  sound 
appear  to  come  from  a  point  some  distance  beyond  the  physical  limits  of  the  loud 
speakers  themselves.  This,  to  the  best  of  our  knowledge,  has  never  been  accom- 
plished before  and  is  a  feature  that  should  open  up  a  new  wealth  of  dramatic 
possibilities  through  the  increase  in  editing  latitude  of  which  we  spoke.  It  is 
particularly  significant  that  this  effect  can  be  produced  not  only  by  a  binaural 
(two-channel)  input  but  also  by  a  monaural  input  of  the  everyday  variety,  as 
has  just  been  demonstrated.  Another  important  fact  is  that  the  point  from  which 
the  sound  appears  to  emanate  can  be  shifted  about  in  space  at  the  will  of  an 
operator  even  though  the  original  record  is  of  the  constant-quality  conventional 
monaural  type.  It  is  now  possible  to  provide  by  this  method  variably  controll- 
able perspective  in  binaural  release-prints  from  any  conventional  sound-track 
as  a  source. 

MR.  GOLDSMITH:    Did  the  same  sound  come  from  each  loud  speaker? 

MR.  OFFENHAUSER:  There  are  two  different  loud  speakers,  one  for  each 
output  channel. 

MR.  GOLDSMITH:  And  you  control  the  amplitude  and  phase  relationships 
between  them? 

MR.  OFFENHAUSER:  The  turntable  over  at  the  left  feeds  into  our  apparatus 
at  the  rear  of  the  room.  From  the  output  of  the  apparatus,  the  system  becomes 


150  W.  H.  OFFENHAUSER,  JR.,  AND  J.  J.  ISRAEL  [J.  s.  M.  P.  E. 

what  might  be  termed  a  conventional  binaural  system.  The  particular  applica- 
tion of  the  quasi-binaural  system  that  we  have  just  demonstrated  is  to  provide 
perspective  in  binaural  release-prints  in  sound  that  had  no  such  perspective  as 
originally  recorded. 

MR.  GOLDSMITH:  You  take  monaural  sound  and  bring  it  in  modified  form  to 
two  separate  speakers  which  play  it  back  in  binaural  form. 

MR.  OFFENHAUSER:  That  is  correct.  The  system  also  permits  the  use  of  a 
conventional  binaural  input  which  produces  other  desirable  effects  not  attain- 
able with  conventional  equipment. 

MR.  CARVER:      Was  someone  varying  the  effects  while  we  were  listening? 

MR.  OFFENHAUSER:  Mr.  Israel,  at  the  back  of  the  room.  He  was  manipu- 
lating the  controls  which  produce  the  effects. 

MR.  WOLF:    What  is  the  relation  between  the  two  acoustic  sources? 

MR.  OFFENHAUSER:  As  mentioned  in  the  paper,  there  appears  to  be  a  differ- 
ence of  opinion  as  to  whether  the  effects  are  a  result  of  phase-difference  or  ampli- 
tude-difference or  both. 

We  can  offer  this  as  a  clue  to  the  answer  to  the  question.  In  certain  tests 
that  we  conducted  with  sounds  of  complex  wave-form,  we  connected  the  vertical 
plates  of  a  cathode-ray  tube  across  the  loud  speaker  leads  of  one  channel  and 
the  horizontal  plates  of  the  tube  across  the  loud  speaker  leads  of  the  other  channel, 
and  then  manipulated  our  controls  as  we  did  in  the  present,  demonstration.  The 
trace  on  the  tube  screen  was  a  straight  line  which  took  different  angular  positions 
on  the  screen  as  our  controls  were  manipulated.  We  obtained  substantially  the 
same  effects  demonstrated  today,  and  throughout  the  variation  the  trace  on  the 
screen  remained  a  straight  line.  Phase-difference  would  thus  appear  to  be  quite 
important. 

MR.  GOLDSMITH:  You  are  asking  that  we  be  aural  "guinea  pigs,"  and  you 
want  to  know  whether  as  such  we  got  the  effect  of  distance  and  of  people  turning 
their  heads,  or  moving  away,  or  moving  forward,  or  turning  toward  a  reflecting 
wall,  and  the  like.  It  would  be  interesting  to  have  the  members  of  the  audience 
express  their  viewpoints,  as  to  whether  they  got  the  impression  of  people  moving 
about  and  turning  their  heads  away  from  the  audience,  and  whether  the  effect 
approximated  the  binaural  action. 

MR.  CUTHBERT:  The  effect  was  very  noticeable.  There  is  one  point,  how- 
ever: How  much  of  the  effect  was  actually  due  to  the  directional  quality  of  the 
speakers? 

MR.  OFFENHAUSER:  Possibly  the  best  answer  to  that  question  is  in  our  past 
experience.  We  made  quite  a  number  of  extensive  tests  with  all  sorts  of  speakers, 
and  as  a  result  we  consider  it  reasonable  to  expect  that  these  binaural  effects 
can  be  reproduced  effectively  if  each  of  the  loud  speakers  used  will  cover  the 
area  properly  as  a  monaural  speaker  in  the  conventional  manner. 

MR.  GOLDSMITH:  You  mean,  then,  that  the  two  speakers  will  operate  satis- 
factorily in  the  overlapped  areas.  That  would  create  an  acoustic  pattern  which 
you  are  shifting  about.  I  noticed  a  definite  change  in  acoustic  quality  and  a 
decided  movement  of  the  voice.  I  had  the  feeling  that  the  control  was  some- 
what overdone  in  two  or  three  places.  There  were  times  when  it  was  handled 
with  more  care  and  feeling.  I  noticed  some  effects  I  thought  were  significant, 
but  there  were  parts,  of  course,  where  you  were  just  exaggerating  the  effects. 


Feb.,  1939]  BlNAURAL  RECORDING  151 

MR.  OFFENHAUSER:  There  is  one  problem  that  always  arises  in  connection 
with  any  binaural  sound  transmission,  and  that  is  the  orientation  of  the  listener 
with  respect  to  the  sound  being  transmitted.  We  will  admit  that  we  tried  it  out 
on  you.  Since  there  was  no  picture,  we  felt  that  it  would  be  a  good  test,  because 
if  we  could  produce  the  illusion  with  sound  alone  without  the  very  material  aid 
that  a  picture  gives,  there  would  be  no  question  that  the  illusion  produced  by  the 
combination  of  picture  and  sound  would  be  heightened  to  the  point  where  it 
would  be  quite  satisfactory. 

MR.  WOLF:    There  was  definitely  no  lag  between  the  two  channels. 

MR.  OFFENHAUSER:  We  have  only  one  input  channel.  There  is  no  electrical 
time  lag  in  the  system. 

MR.  WOLF:    It  appears  to  be  nothing  more  than  a  phase  change. 

MR.  OFFENHAUSER:  Not  quite.  As  we  have  explained  before,  it  is  more 
like  a  combination  of  both  phase  and  amplitude  control. 

MR.  MITCHELL:  What  difference  would  it  make  if  the  loud  speakers  were 
separated,  one  on  each  side  of  the  room.  I  think  it  is  customary  to  assume  that 
binaural  speakers  will  be  placed  quite  widely  separated. 

MR.  OFFENHAUSER:  While  wide  speaker  placement  has  been  found  by  cut- 
and-try  methods  to  be  most  satisfactory  in  conventional  binaural  systems,  our 
results  seem  to  indicate  no  such  limitation  for  this  system;  in  fact,  speakers 
placed  side  by  side,  as  we  have  placed  them  here,  produce  the  most,  shall  we  say, 
startling  effects? 

MR.  GOLDSMITH:    Dramatic  effects? 

MR.  OFFENHAUSER:  That  is  a  better  description.  Of  course,  in  the  initial 
public  demonstration  of  a  system  of  this  sort  we  must  try  to  obtain  exaggerated 
effects  in  order  that  you  may  leave  with  some  lasting  impressions.  If  we  were 
to  show  this  in  its  true  aspect  as  it  would  be  shown  with  a  picture,  the  sound 
demonstration  without  the  picture  would  not  be  as  impressive.  That  is  entirely 
psychological,  as  you  are  aware. 

MR.  MACNAIR:  This  paper,  on  two  channel  reproduction,  emphasizes  again 
before  the  Society  that  our  normal  equipment  is  a  one-channel  system  from 
the  sound  stage  through  to  the  theater,  and  as  such  has  certain  inherent  limita- 
tions. 

I  should  like  to  make  one  appeal,  which  may  seem  to  many  of  you  here  to  be 
a  purist's  appeal.  This  is  on  the  use  of  this  word  "binaural."  Sooner  or  later, 
if  this  type  of  reproduction  becomes  common,  the  Standards  Committee  is  going 
to  be  up  against  the  choice  of  standards  and  nomenclature  and  it  will  ease  their 
work  if  we  pay  attention  to  the  exact  meaning  of  the  word.  Normally  a  bin- 
aural  system  refers  to  a  system  in  which  the  sound  is  picked  up  by  a  dummy 
or  something  similar  with  a  microphone  to  replace  each  ear.  The  left  ear  of  the 
dummy  hears  sounds  that  would  be  heard  by  the  left  ear  of  an  observer  in  that 
position,  and  the  right  ear  of  the  dummy  would  hear  the  sound  that  would  be 
heard  by  the  right  ear  of  an  observer  in  that  position. 

The  word  "binaural"  has  been  used  to  refer  to  that  type  of  system  which,  in 
a  very  strict  way,  transfers  those  two  channels  to  a  right  and  left  ear  of  the 
ultimate  observer.  The  characteristic  of  that  system  is  that  the  sound  that  is 
heard  by  the  left  ear  of  the  dummy  is  transmitted  only  to  the  left  ear  of  the 


152  W.  H.  OFFENHAUSER,  JR.,  AND  J.  J.  ISRAEL  [J.  S.  M.  P.  E. 

observer,  and  the  sound  that  is  received  by  the  right  ear  of  the  dummy  is  trans- 
ferred only  to  the  right  ear  of  the  observer. 

I  think  it  would  make  our  discussion  easier  if  the  word  "binaural"  were  left 
for  that  unique  and  accurately  specified  system,  and  that  these  other  systems, 
which  give  a  sense  of  motion  of  the  sound,  were  referred  to  by  some  other  name. 

MR.  OFFENHAUSER:  In  all  the  earlier  patents  and  scientific  publications  such 
as  in  the  Physical  Review,  the  term  "binaural"  seems  to  have  been  adopted  as 
descriptively  generic,  and  we  have  used  the  word  in  that  sense  as  being  descrip- 
tive of  certain  effects  rather  than  of  certain  systems.  The  dictionary,  while  not 
particularly  precise,  seems  to  lean  in  the  same  direction  (binaural — having  or 
relating  to  two  ears;  involving  the  use  of  both  ears:  Merriam  Webster  New 
International,  1936}. 

MR.  MACNAIR:  I  wanted  to  point  out  that  from  the  physical  and  engineering 
points  of  view  there  are  more  than  one  kind  of  system.  We  ought  to  admit  from 
the  outset  that  there  are  basically  different  kinds  of  systems,  and  therefore  they 
should  be  referred  to  with  different  words. 

MR.  OFFENHAUSER:  Until  the  present  time  it  has  hardly  been  possible  to 
classify  such  systems  since  a  particular  effect  produced  in  any  of  the  previous 
systems  usually  had  as  an  inseparable  accompaniment  other  effects,  usually 
undesirable,  which  were  not  independently  controllable  by  any  suitable  means. 

There  still  remains  the  problem  of  properly  classifying  the  effects  that  we  do 
consider  desirable.  It  will  require  a  consensus  of  opinion  to  determine  suitable 
and  proper  classifications,  and  in  order  that  that  may  be  accomplished,  there  is 
required  a  study  of  the  individual  effects  of  themselves.  One  example  is  room 
echo,  or  effect,  or  reverberation. 

Let  us  suppose  that  this  room  is  an  acoustically  simple  room,  without  columns 
but  with  very  hard,  flat  walls,  and  a  hard  flat  ceiling.  When  a  person  talks  in 
such  a  room,  a  listener  gets  one  impression  of  the  room's  size.  If  we  block  off 
one-half  of  the  room,  as  with  a  plaster  block  wall,  the  listener  has  another  im- 
pression merely  because  the  size  and  shape  of  the  room  have  changed. 

We  do  not  at  the  present  time  have  technical  terms  to  describe  accurately  the 
effects  that  we  feel.  Unless  the  effects  are  segregated,  it  would  seem  to  me  that 
we  are  going  to  have  difficulty  with  any  classifications  that  we  may  use  in  analysis. 

Let  us  again  consider  our  acoustically  simple  room.  We  shall  now  modify  it 
by  using  some  absorption  material  uniformly  over  all  the  exposed  surfaces. 
With  listener  and  speaker  in  the  same  places  as  in  the  "hard"  room,  the  effects 
will  be  different.  The  effect  produced  when  the  speaker  is  walking  about  in  the 
"hard"  room  is  certainly  different  from  the  effect  produced  when  he  is  walking 
about  in  the  treated  room. 

If  the  room  is  then  further  modified  so  that  it  becomes  a  room  such  as  this  is 
acoustically,  with  its  columns  and  its  windows  and  its  partially  treated  surfaces, 
the  effect  produced  when  the  speaker  walks  about  in  such  a  room  is  certainly 
different  from  any  of  the  other  conditions  described. 

MR.  GOLDSMITH:  Yes,  you  have  mobility  of  the  speaker  and  you  have 
sonority  of  the  surroundings,  and  you  mix  the  sonority  of  the  surroundings  and 
the  mobility  of  the  sound  source  and  get  a  wide  variety  of  acoustic  effects. 

MR.  OFFENHAUSER:  With  this  arrangement,  it  is  possible  to  simulate  many 
of  these  effects  in  reproduction  when  the  sound-source  or  sources  actually  remain 


Feb.,  1939]  BlNAURAL  RECORDING  153 

stationary  with  respect  to  the  microphone.  It  is  even  possible  to  take  a  small 
orchestra,  set  up  compactly  for  ordinary  pick  up  and,  by  the  manipulation  of 
our  controls,  cause  the  reproduced  orchestra  to  appear  to  spread  out  over  a  large 
area  of,  let  us  say,  1000  square-feet,  or  to  contract  into  the  area  in  which  such 
a  group  of  musicians  would  ordinarily  perform.  In  addition  to  mobility  and 
sonority  we  also  seem  to  have  a  characteristic  of  the  apparent  source  itself  which 
we  can  call  source  size.  This,  too,  is  a  variable. 

There  is  a  decided  advantage  in  this  particular  type  of  reverberation  effect 
inasmuch  as  it  is  synthetic  and  is  not  dependent  upon  the  limitations  of  physical 
surroundings. 

MR.  CRABTREE:  It  was  my  feeling  that  you  kept  switching  from  one  speaker 
to  the  other.  Would  not  the  test  have  been  better  if  you  had  covered  up  the 
loud  speakers? 

MR.  OFFENHAUSER:  That  is  a  good  point.  We  had  considered  darkening  the 
room  and  projecting  a  picture  or  some  kind  of  visual  color  pattern  but  decided 
against  doing  so  in  order  that  we  might  demonstrate  the  system  to  you  under 
the  worst  possible  psychological  conditions.  If  we  obtain  results  under  these 
conditions,  we  can  consider  this  an  acid  test. 

MR.  GRIFFIN  :  Is  the  particular  type  of  speaker  that  you  are  using  of  necessity 
a  part  of  the  system? 

MR.  OFFENHAUSER:  Definitely  not.  We  have  used  flat  baffle  speakers, 
directional  horn  speakers,  and  all  sorts  of  other  speakers,  and  can,  to  a  varying 
degree,  produce  the  effects  with  commercial  speakers.  Loud  speakers  with  good 
high-frequency  response  are  preferable,  however. 

MR.  ENGL:  As  I  understand  it,  there  are  two  separate  electrical  channels 
connected  to  the  two  loud  speakers.  It  might  be  an  interesting  experiment  to 
connect  both  channels  to  one  loud  speaker  instead  of  two  loud  speakers.  As 
far  as  phase-differences  are  concerned  the  result  on  the  audience  should  be  ex- 
pected to  be  the  same. 

MR.  OFFENHAUSER:  If  they  were  connected  to  the  same  loud  speaker  mecha- 
nism, you  would  get  an  effect  similar  to  mixing. 

MR.  ENGL:     I  mean  to  use  the  same  mixing  system. 

MR.  OFFENHAUSER:     It  is  impracticable  to  try  it  now. 

MR.  KELLOGG:  I  should  like  to  second  Mr.  MacNair's  remarks  and  make 
a  plea  for  reserving  the  term  "binaural"  for  a  really  completely  separated  channel 
between  two  microphones.  I  believe  the  use  of  that  term  in  patents  is  not  as 
good  a  criterion  as  to  what  is  the  best  usage  of  language  as  the  scientific  writings, 
because  those  who  write  patents  may  have  excellent  ideas  and  a  good  deal  of 
ingenuity,  but  as  a  class  they  are  not  as  careful  probably  as  those  who  come 
along  afterward  and  reduce  things  to  careful  analysis  and  classification.  I 
believe  that  "binaural"  is  usually  used  by  scientific  writers  in  the  sense  of  sepa- 
rated channels. 

MR.  OFFENHAUSER:  We  have  substantially  separated  output  channels  here, 
but  our  input,  as  we  have  described,  is  a  single  channel. 

MR.  KELLOGG:  Still  it  is  not  a  binaural  channel  in  the  sense  I  have  just 
described. 

MR.  OFFENHAUSER:  That  is  correct;  we  call  it  quasi-binaural.  We  would 
like  to  call  your  attention  to  the  fact,  however,  that  this  system  is  by  no  means 


154  W.  H.  OFFENHAUSER,  JR.,  AND  J.  J.  ISRAEL  [j.  s.  M.  p.  E. 

limited  to  monaural  input  and  may  be  used  with  even  better  results  with  binaural 
microphone  input.  What  we  have  demonstrated  is  the  quasi-binaural  system. 

MR.  KELLOGG:  That  may  be  all  right.  It  may  be  of  interest  to  you  that  a 
number  of  years  ago  I  witnessed  a  number  of  tests  at  our  laboratories  in  Camden, 
in  which  we  resorted  to  changes  in  phase  and  changes  in  intensity  to  shift  sound 
apparently  from  one  source  to  another,  and  it  appeared  that  either  an  advance 
in  phase  or  an  increase  in  intensity  would  serve  to  shift  the  sound  from  one  source 
to  another.  Within  moderate  limits,  phase  has  preference,  for  if  the  intensity  is 
pretty  close,  the  one  from  which  the  sound  reaches  you  first  is  what  you  consider 
to  be  the  real  source.  However,  with  further  increase  in  the  difference  in  level, 
you  presently  reach  a  point  where  the  louder  one  seems  to  be  the  source.  This 
is  when  it  almost  completely  subdues  the  other.  Apparently  our  ears  judge,  so 
far  as  possible,  by  where  the  first  impulse  of  sound  comes  from. 

MR.  DAVEE:  There  are  one  or  two  points  in  this  system  that  we  should  not 
overlook.  The  first  was  mentioned  by  Mr.  Offenhauser  and  I  would  like  to 
emphasize  it :  the  possibility  of  moving  the  sound  outside  the  limits  of  the  loud 
speakers. 

Second,  if  each  of  you  should  walk  about  the  room  as  we  did  last  night,  you 
would  realize  the  rather  complete  coverage  of  the  effect  in  this  auditorium.  If 
you  have  ever  worked  with  binaural  systems  before,  you  know  that  the  effect 
is  usually  quite  disappointing  when  you  get  very  far  from  the  loud  speakers. 
With  this  system,  from  my  observation  in  walking  about  the  room,  the  effect  is 
very  startling.  One  can  walk  about  the  room  to  his  heart's  content  and  the 
binaural  effect  is  still  there.  The  word  "binaural"  appears  to  me  to  be  an  ad- 
jective describing  an  effect  rather  than  the  name  of  a  system. 

MR.  WOLF:  This  system  is  definitely  not  a  binaural  system  in  the  way  we 
are  accustomed  to  thinking  of  it.  With  regard  to  the  phase  relations,  I  thought 
it  was  an  accepted  fact  that  they  did  not  make  any  difference  in  the  acoustic  or 
subjective  effect. 

MR.  ENGL:  Perhaps  I  should  not  have  used  the  term  phase-difference,  but 
rather  time-difference.  In  other  words,  the  effect  here  in  question  probably 
depends  upon  time  differences.  My  idea  was  that  similar  effects  could  be  pro- 
duced with  one  loud  speaker  because  you  can,  of  course,  get  two  sound  waves 
from  one  loud  speaker  with  the  desired  time-difference  between  them.  With 
regard  to  the  expression  "binaural,"  I  think  one  might  propose  the  term  "stereo- 
phonic." 

MR.  GOLDSMITH:  In  connection  with  phase  delay,  what  is  generally  under- 
stood is  that  if  the  various  frequency  components  of  a  given  sound  are  altered  in 
phase,  the  effect  is  not  noticeably  detectable.  But  if  you  have  two  complete 
replicas  of  a  given  sound,  with  all  the  components  shifted  bodily,  a  new  effect 
of  an  echo  occurs. 

MR.  MITCHELL:  What  is  the  effect  of  the  reverberation  of  the  auditorium, 
comparing  a  live  room  with  a  comparatively  dead  room  with  the  present 
set-up? 

MR.  OFFENHAUSER:  With  this  type  of  system,  we  can  go  into  a  room  where 
reproduction  is  practically  impossible  with  commercial  conventional  equipment 
and  where  even  direct  speech  from  a  live  speaker  is  unsatisfactory,  and  obtain 
improved  intelligibility ;  even  better  in  certain  cases  than  the  live  speaker  himself. 


Feb.,  1939]  BlNAURAL  RECORDING  155 

There  is  one  other  effect  we  have  found  that  seems  to  be  important  and  we  are 
going  to  investigate  it  much  further  than  we  have  so  far.  In  the  application  to 
public  address  systems,  we  were  able  to  adjust  our  apparatus  in  such  a  manner 
that  we  could  increase  the  sound  level  in  an  auditorium  at  the  feedback  point 
by  some  6  or  8  db.  At  the  present  time  we  attribute  it  to  a  form  of  negative 
reaction  effect  between  the  sound  projected  by  the  loud  speakers  and  that  picked 
up  by  the  microphone  or  microphones. 

MR.  GOLDSMITH:  That  is  probably  acoustic  degeneration  between  loud 
speakers  and  microphones. 

MR.  OFFENHAUSER:     Quite  likely. 


COORDINATING  ACOUSTICS  AND  ARCHITECTURE  IN 
THE  DESIGN  OF  THE  MOTION  PICTURE  THEATER* 

C.  C.  POTWIN**  AND  B.  SCHLANGERf 

Summary. — In  past  practice,  acoustics  has  been  overlooked  as  a  function  of  the 
architectural  planning  of  motion  picture  theater  structures.  The  need  for  and  the 
extensive  use  of  sound-absorbing  devices  in  existing  buildings  have  led  to  reliance 
upon  corrective  methods  in  the  planning  of  new  designs. 

The  constructive  approach  to  the  solution  of  acoustical  problems  in  new  design  is 
achieved  only  through  proper  determination  of  the  basic  proportions,  cubic-foot 
volume  per  seat,  and  detailed  form  of  the  auditorium  structure. 

The  purpose  of  this  paper  is  to  show  that  acoustics  can  be  coordinated  construc- 
tively with  the  other  primary  functions  of  theater  planning  to  develop  a  more  efficient 
and  more  economical  design  and  one  that  truly  expresses  modern  and  creative  archi- 
tecture. 

A  critical  distinction  of  modern  architecture  from  prior  technic 
lies  in  a  more  candid  evaluation  of  function.  Because  of  the  persis- 
tence of  superficial  ornamentation,  however,  this  fundamental  is  fre- 
quently lost  in  our  conception  of  what  really  constitutes  modern 
design.  We  need  not  here  go  into  the  history  of  the  modern  approach 
to  architectonics ;  suffice  it  to  point  out  that  it  assuredly  did  not  have 
its  real  origin  in  ornamentation.  Instead,  it  began  as  a  method  of 
using  our  standard  and  new  materials  for  the  creation  of  buildings 
better  suited  to  our  need  of  them. 

Despite  widespread  digressions,  which  continue  to  exalt  the  purely 
aesthetic  evaluations,  modernism  in  architecture  is  the  offspring  of 
science,  not  of  fine  art.  Modern  science  has  revised  our  approach  to 
most  things.  Giving  us  knowledge  of  a  thousand  venerable  mysteries, 
it  has  discouraged  circumlocution  and  falsity  in  our  expressions  gen- 
erally. To  be  really  modern  in-  architecture  is  to  go  straight  to  the  purpose 
of  a  building,  and  to  develop  it  in  plan  and  structure  according  to  an  honest 
acceptance  of  that  purpose,  providing  in  the  forms  and  devices  that 
serve  it  a  beauty  that  is  inherent. 

*  Presented  at  the  1938  Fall  Meeting  at  Detroit,  Mich.;   received  December 
12,  1938. 

**  Electrical  Research  Products,  Inc.,  New  York,  N.  Y. 
t  Theater  Architect,  New  York,  N.  Y. 

156 


ACOUSTICS  AND  ARCHITECTURE  157 

When  the  ideal  of  functional  efficiency  and  proper  aesthetic  quality 
becomes  our  guide  in  designing  a  motion  picture  theater,  the  precepts 
of  sound,  light,  and  vision  supply  the  basis  for  fundamental  planning. 
Fortunately,  the  insistence  upon  such  a  basis  has  lost  much  of  its 
radicalism,  for  which  thanks  are  in  a  great  measure  due  this  Society. 
Through  the  efforts  of  the  Projection  Practice  Committee1  and  of 
other  groups  and  individuals,  definite  advances  have  been  made  in 
many  engineering  phases  of  theater  planning.  Correct  vision,  without 
obstruction,  has  been  provided  in  a  number  of  new  designs,  although 
much  additional  work  is  required  in  this  field.  Lighting  in  the  audi- 
torium, coincidental  with  the  picture  presentation,  is  now  being 
given  more  study.  Other  provisions  for  the  patrons'  comfort  and 
enjoyment,  such  as  air-conditioning,  proper  seating,  and  general 
arrangement  and  appointments  are  also  receiving  much  more  con- 
sideration. Unfortunately,  however,  practically  no  attention  has 
been  given  in  previous  design  practice  to  the  relationship  that  exists 
between  acoustics  and  the  fundamental  plan  of  the  theater.  This 
does  not  mean  that  the  acoustical  problem  has  received  no  attention. 
On  the  contrary,  it  has  often  received  thoughtful  treatment,  but  more 
specifically  from  what  might  be  termed  a  corrective  rather  than  a 
constructive  point  of  view.  In  other  words,  the  usual  procedure  has 
been  first  to  plan  the  theater  from  all  other  aspects  and,  in  many 
cases,  even  to  go  so  far  as  to  determine  the  decorative  treatment  of 
the  interior,  before  considering  acoustics.  Then,  as  the  final  step  (in 
cases  where  the  subject  of  acoustics  is  given  consideration  in  planning) 
sound-absorbing  materials  are  selected  with  a  view  to  correcting  or 
compensating  for  acoustical  deficiencies  in  the  design — definite  as- 
surance being  made  first  that  these  materials  will  fit  a  predetermined 
decorative  scheme. 

This  corrective  approach  to  the  solution  of  acoustical  problems  has 
become  common  practice  because:  (1)  the  preliminary  basic  form  of 
auditoriums  has  not  been  planned  for  best  acoustics;  (2)  the  total 
volume  of  the  auditorium  structure  has  not  been  held  down  so  as 
to  fall  within  desirable  limits,  as  it  might  have  been  in  many  new 
designs;  and  (3)  the  past  tendency  to  follow  tradition  in  architectural 
design  practice  has  usually  made  it  mandatory  to  utilize  corrective 
methods.  There  is  one  other  reason  that  should  not  be  overlooked.  At 
the  time  speech  and  music  were  added  to  films,  sound-absorbing  de- 
vices were  required  for  many  existing  structures  having  poor  acousti- 
cal conditions.  Perhaps  from  force  of  habit,  reliance  upon  these  de- 


158  C.  C.  POTWIN  AND  B.  SCHLANGER  [J.  S.  M.  P.  E. 

vices  has  extended  into  the  planning  of  new  theaters.  As  a  result, 
basic  acoustical  design  has  been  overlooked  as  one  of  the  sources  of 
effective  architecture. 

It  is  not  the  purpose  of  this  paper  to  discourage  the  use  of  acoustical 
materials  for  the  treatment  of  motion  picture  theaters.  Such  ma- 
terials may  be  required  for  the  following  special  cases:  namely,  for 
new  theaters  having  very  large  seating  capacities ;  for  existing  theaters 
having  fixed  forms  that  produce  objectionable  sound  reflections ;  and 
for  both  new  and  existing  structures  having  excessive  cubic-foot 
volumes  per  seat. 

It  is  proposed  to  show,  however,  that  more  efficient  and  more 
economical  theater  structures  can  be  built  when  basic  acoustical  re- 
quirements are  coordinated  with  the  other  primary  functions  of 
theater  planning.  There  are  four  outstanding  reasons  why  this  con- 
structive approach  will  produce  more  successful  results:  (1)  today 
the  various  elements  affecting  the  control  of  sound  in  a  design  can  be 
studied  initially  and  planned  correctly,  with  the  result  that  very  little 
or  no  acoustical  material  need  be  provided  for  the  wall  or  ceiling  sur- 
faces; (2)  minimum  surface  treatment  makes  for  better  acoustics, 
because  (a)  the  proper  character  of  reverberation  and  absorption  can 
be  assured,  and  (b)  less  critical  conditions  need  finally  be  met  in 
balancing  the  frequency  absorption  characteristic  of  the  theater;  (3) 
in  cases  where  little  or  no  acoustical  material  is  required,  the  architect 
is  at  liberty  to  use  ordinary,  every-day  materials,  thereby  making  for 
greater  flexibility  in  design;  and  (4)  when  a  theater  is  efficiently 
planned  in  this  way,  substantial  economies  can  be  realized  not  only  in 
acoustical  treatment  but  also  in  other  phases  of  theater  planning. 

From  the  architectural  standpoint,  planning  for  proper  acoustical 
conditions  in  the  initial  design  stage  does  not  preclude  the  ability 
to  obtain  pleasing  forms  or  surface  finishes  for  the  auditorium.  Ac- 
tually, creative  design  is  more  readily  inspired  by  this  approach. 

ACOUSTICAL  REQUIREMENTS 

Two  fundamental  factors  that  must  be  considered  as  the  first  step 
in  the  functional  acoustic  planning  of  a  motion  picture  theater  are: 
(1)  the  preliminary  outline  or  basic  form  of  the  auditorium,  establish- 
ing its  proportions  of  length,  width,  and  height;  and  (2)  the  volume 
or  cubical  content  of  the  auditorium  structure  in  its  relationship 
to  the  seating  capacity. 


Feb.,  1939]  ACOUSTICS  AND  ARCHITECTURE  159 

Actual  design  practice  indicates  that  the  most  efficient  control  of 
sound  reflections  and  the  best  distribution  of  sound  energy  can 
usually  be  obtained  in  theater  auditoriums  where  the  ratios  of 
width  to  length  fall  within  the  limits  of  1  :  1.4  and  1  :  2.2  When  the 
the  length  becomes  greater  than  twice  the  width,  difficulties  arise  from 
a  multiplicity  of  sound  reflections  occurring  between  the  side  wall 
surfaces.  When  the  ratio  of  width  to  length  is  less  than  1 :  1.4,  the  re- 
sulting design  becomes  an  unfavorable  one  from  the  standpoints  of 
proper  sound  distribution  and  vision.  Furthermore,  this  design 
creates  an  unusally  large  rear  wall,  which  is  often  a  source  of  objec- 
tionable sound  reflections. 

The  limits  of  ratios  recommended  above  for  the  floor  plan  are  not 
meant  to  suggest  that  a  strictly  rectangular  outline  must  be  adopted 
for  the  fundamental  design.  These  ratios  also  apply  to  the  average 
dimensions  for  the  width  and  length  of  an  irregular  basic  outline, 
such  as  one  having  a  moderate  splay  in  the  preliminary  form  of  the 
side  walls  or  a  generally  nonsymmetrical  arrangement  of  outline 
surfaces. 

Establishing  the  ceiling  height  in  its  most  practicable  and  best 
acoustical  relationship  to  the  horizontal  dimensions  is  the  next 
element  to  be  considered  in  fundamental  planning.  Ceiling  height  is 
very  important  because  it  affects  both  the  proportions  and  the 
structural  volume  of  the  auditorium. 

From  an  architectural  viewpoint,  the  determination  of  ceiling 
height  is  governed  by  three  factors,  namely:  (1)  sight-line  require- 
ments; (2)  width  of  the  light-beam  and  its  projection  angle  to  the 
screen;  and  (3)  the  general  appearance  of  the  auditorium. 

From  the  acoustical  standpoint,  two  other  important  factors  should 
be  included  in  the  determination  of  ceiling  height.  These  are:  (1) 
the  proper  relationship  of  the  ceiling  height  dimensions  to  the  hori- 
zontal proportions;  and  (2)  the  optimum  cubic-foot  volume  per  seat 
required  for  a  given  design. 

A  ceiling-height  ratio  can  not  be  fixed  that  will  be  adaptable  to  all  de- 
signs. The  best  ratio  can  be  established  only  by  a  study  of  the  hori- 
zontal dimensions  and  by  a  preliminary  analysis  of  the  cubic-foot  vol- 
ume requirements  for  the  initial  control  of  reverberation  time  in  a 
given  design. 

Excessive  ceiling  heights  are  commonly  found  today  in  auditorium 
design  practice.  The  large  structural  volume  per  seat  and  the  pro- 
longed reverberation  time  resulting  from  this  acoustical  defect  in 


160  C.  C.  POTWIN  AND  B.  SCHLANGER  [J.  S.  M.  P.  E. 

theater  planning  have  produced  a  need  for  corrective  materials  in 
many  new  designs. 

The  following  study  will  show  the  variations  in  basic  proportions 
and  structural  volume  introduced  in  the  planning  of  a  theater  audi- 
torium seating  900  persons.  This  study  will  also  show  the  reasons 
for  these  variations  and  how  they  can  be  minimized. 

From  an  acoustical  standpoint,  the  structural  volume  of  an  audi- 
torium seating  900  persons  should  lie  between  120  and  130  cubic- 
feet  per  seat.  The  factors  affecting  the  determination  of  these  limits 
are:  (1)  the  optimum  or  best  times  of  reverberation  at  different  fre- 
quencies for  reproduced  sound,3  and  (2)  the  fixed  and  variable  sound- 
absorption  required  to  produce  these  times  of  reverberation. 

Fixed  absorption  means  the  absorption  provided  by  theater  chairs, 
carpets,  and  interior  surfaces  of  standard  furred  construction,  finished 
in  an  ordinary  manner.  Variable  absorption  is  the  absorption  nor- 
mally provided  by  the  audience.  "Optimum"  reverberation  times  are 
usually  based  on  an  audience  condition  approximating  two-thirds  of 
the  house  seating  capacity. 

The  chairs  constitute  the  major  part  of  the  fixed  absorption  for  a 
theater  initially  planned  to  be  acoustically  functional  in  design.  To 
ensure  that  the  variable  absorption,  that  is,  the  audience,  will  not 
effect  a  major  change  in  the  reverberation  time,  it  is  very  important 
that  these  chairs  be  of  an  efficient  upholstered  type.  The  type  com- 
monly used  for  theater  seating  today  is  one  having  a  leather-covered 
spring  bottom  and  a  fully  padded  mohair  or  tapestry-covered  back. 
In  the  standard  20-inch  width  this  chair  has  a  sound  absorption  value 
equivalent  to  more  than  two-thirds  that  of  the  average  person.  The 
use  of  such  a  chair  has  been  assumed  for  the  following  study. 

VISUAL  REQUIREMENTS 

From  the  standpoint  of  vision  the  proportions  and  cubic-foot  vol- 
umes of  auditoriums  vary  for  three  different  reasons.  These  are : 

(1}  The  type  of  seating  arrangement  used  in  the  horizontal  dimension  of  the 
auditorium  determined  by  the  number  of  seats  and  aisles  to  be  arranged  across 
the  width  of  the  auditorium.  Types  in  common  use  are  (a)  14  seats  with  a  wall 
aisle  on  either  side  of  the  seating  area;  (6)  2  aisles  separating  three  banks  of 
seats,  the  middle  bank  being  14  seats  wide  and  the  side  banks  against  the  walls 
being  7  seats  wide;  (c)  2  banks  of  14  seats  each  separated  by  a  single  aisle  having, 
in  addition,  2  wall  aisles;  and  (d}  three  banks  of  14  seats,  each  separated  by  two 
aisles  having,  in  addition,  two  wall  aisles. 


Feb.,  1939] 


ACOUSTICS  AND  ARCHITECTURE 


161 


(2)  The  type  of  seating  arrangement  employed  in  the  vertical  dimension  of 
the  auditorium,  such  as  (a)  all  seats  on  a  single  level;  (&)  a  single  general  level 
of  seats  with  the  rear  portion  of  the  seating  area  somewhat  elevated,  with  a 
cross-over  separating  the  raised  portion  from  the  remainder,  this  type  being 
commonly  called  a  stadium  design;  (c)  a  modified  stadium  design  in  which  the 
raised  portion  of  seats  in  the  rear  is  placed  high  enough  so  as  to  cover  the  cross- 
over mentioned  in  the  b  type;  and  (d)  use  of  upper  seating  levels  partially  over- 
hanging the  seating  area  of  the  level  below.4 

(5)  The  inclination  or  inclinations  of  the  floor  for  the  main  or  orchestra  level, 
such  as  (a)  a  steep  downward  inclination  toward  the  screen;  (6)  a  modified 
downward  inclination  toward  the  screen;  (c)  a  modified  downward  and  then  up- 
ward inclination  toward  the  screen;  and  (d)  only  a  slight  downward  and  then 
upward  inclination  toward  the  screen. 


VOLUME  •   l*O.o.K.   FES    SCAT 


VOLUME     •      136*  e*,,  »t    PI*   SIM 


E 

120i   cu  r,    fft    SEAT 


FIG.  1.     Longitudinal  sections  for  five  values  of  cu.-ft.  volumes  per  seat. 


The  degree  of  inclination  of  the  floor  in  all  cases  is  affected  by  the 
manner  in  which  the  seats  are  placed  behind  each  other.  When  these 
seats  are  arranged  in  staggered  fashion  so  that  the  center  of  any  seat 
is  always  directly  behind  the  dividing  armblock  of  the  preceding 
seat,  the  inclination  of  the  floor  may  be  reduced  by  almost  one-half. 
Variations  in  floor  inclination  directly  affect  the  height  and  inclina- 
tion of  the  upper  seating  levels  and  the  placing  of  the  projection 
room,  thereby  the  total  height  of  the  auditorium. 


162  C.  C.  POTWIN  AND  B.  SCHLANGER  [J.  S.  M.  P.  E. 

ANALYSIS  OF  TYPICAL  DESIGNS 

The  study  presented  in  Fig.  1  has  been  made  to  investigate  which 
of  the  various  possible  auditorium  designs  would  prove  most  efficient 
from  both  the  visual  and  auditory  standpoints,  keeping  in  mind  that 
economy  in  construction  and  architectural  appearance  are  also  im- 
portant guiding  factors.  An  auditorium  seating  900  persons  was 
selected  for  this  study  because  this  capacity  has  been  found  to  answer 
the  needs  of  the  average  motion  picture  theater  and  because  a  large 
proportion  of  the  theaters  now  being  erected  approximate  this 
capacity.5 

It  is  interesting  to  note  in  the  five  theater  designs  shown  that  the 
structural  volumes  of  the  auditoriums  vary  from  108,450  to  134,100 
cu.-ft.  Furthermore,  the  required  screen  width  varies  from  a  mini- 
mum of  lQl/z  ft.  to  a  maximum  of  22  ft.  due  to  the  increases  in  maxi- 
mum viewing  distance.  Such  variations  in  design  are  not  justified 
when  the  seating  capacity  is  the  same  in  all  cases,  as  it  is  in  all  the 
designs  here  shown. 

TABLE  I 

Cu.-Ft.  of  Max.  Viewing 

Design  Vol.  per  Seat  Distance  Screen  Width 

A  151  104ft.  20ft. 

B  148  119  22 

C  130  105  20 

D  136V2  89  16       6  in. 

£  1201/*  89  16       6 

Ceiling  heights  have  been  kept  down  to  a  minimum  in  all  five  de- 
signs, and  the  seating  arrangement  in  the  horizontal  dimension  is  the 
same  in  all  cases,  using  the  type  having  two  aisles  separating  three 
banks  of  seats,  the  middle  bank  being  fourteen  seats  wide  and  the 
side  banks  against  the  walls  being  seven  seats  wide.  This  arrange- 
ment in  the  horizontal  dimension  was  selected  for  these  studies  be- 
cause it  is  the  most  efficient  plan  for  capacities  of  approximately  900 
seats,  requiring  the  least  amount  of  aisle  space  and  thereby  effec- 
tively reducing  the  total  volume  for  all  the  design  studies. 

Table  I  shows  the  resulting  characteristics  of  the  five  designs  shown 
in  Fig.  1. 

In  the  design  types  A,  B,  and  D  shown  in  Fig.  1,  the  slope  of  the 
orchestra  floor  is  approximately  what  has  been  commonly  used  in 
past  practice.  This  slope  ordinarily  does  not  afford  a  sufficiently 
unobstructed  vision  of  the  screen.  To  correct  this  defect,  an  increase 


Feb.,  1939] 


ACOUSTICS  AND  ARCHITECTURE 


163 


in  the  slope  in  these  types  would  result  in  a  slight  increase  in  the  total 
volume  where  the  volume  is  already  excessive.  However,  visibility 
could  also  be  improved  in  these  designs  by  the  use  of  a  stagger  system 
of  seating  which  would  effect  a  slight  reduction  in  the  total  volume. 
This  reduction  would  benefit  design  D  only.  Designs  A  and  B  would 
still  provide  excessive  volume. 

In  designs  C  and  E  the  staggered  system  of  seating  has  been  used 
for  the  lower  seating  areas.  However,  a  non-staggered  plan  could  also 
be  used  which  would  result  in  a  floor  that  would  pitch  downward 
toward  the  screen  approximately  one  foot  more  and  pitch  upward 
toward  the  screen  in  the  front  half  of  the  floor  to  a  point  about  level 
with  the  floor  behind  the  seats  farthest  from  the  screen.  The  volume 


20 

/    ^ 

^ 

\ 

•NtHA 
PER  S 

45  cu 

:E  VA 

'Ml 

j 

\ 

NET 

.OE 

\ 

13 

1 

\ 

I 

\ 

1 

^ 

1 

^ 

•^ 

f 

T 

\ 

7 

\ 

j 

\^ 

^"^ 

^^_ 

\ 

N 

S 

^S* 

•CLOW  -  K 

O         1 

0       i; 

0       U 

0         M 

0       U 

0         M 

0         1 

0        14 

O          M 

0       2 

>0-ABOVE 

VOLUME -CUBIC  FEET  PER  SEAT 

FIG.  2.     Volume  of  100  theaters  seating  800-1000  persons. 

added  to  designs  C  and  E  due  to  using  a  non-staggered  plan  would 
raise  the  total  volume  only  a  negligible  amount. 

The  designs  shown  here  have  horizontal  shape  ratios  varying  from 
1  :  1.65  in  designs  D  and  E  to  1  :  2  in  design  B.  Although  all  these 
horizontal  ratios  come  within  the  acoustical  limits  heretofore  recom- 
mended, it  is  particularly  significant  from  the  analyses  that  the  proper 
vertical  solution  of  a  design  not  only  results  in  good  basic  form  but  in 
most  cases  is  the  primary  factor  determining  the  efficient  control  of 
structural  volume. 

The  data  given  in  Fig.  2  provide  an  interesting  parallel  to  this  study 
and  show  the  variations  in  volume  per  seat  for  one  hundred  theaters 
built  recently,  seating  between  800  and  1000  persons.  Seventy-one 
of  this  number  are  theaters  of  the  balcony  type,  whereas  only  twenty- 


164  C.  C.  POTWIN  AND  B.  SCHLANGER  [J.  S.  M.  P.  E. 

nine  are  single-floor  houses.  It  is  important  to  note  that  the  peak  of 
this  curve  for  volume  per  seat  vs.  the  number  of  cases  lies  between 
140  and  160  cu.-ft.,  and  that  the  net  average  value  for  the  100  cases 
is  145  cu.-ft.  per  seat.  These  data  suggest  that  the  greater  percentage 
of  present-day  theater  structures  have  excessive  volumes  per  seat. 
This  fact  is  in  turn  a  direct  indication  of  the  lack  of  thought  given  to 
coordinating  the  auditory  and  visual  requirements  in  fundamental 
planning,  particularly  since  in  the  majority  of  these  cases  excessive 
volume  was  a  result  of  excessive  ceiling  height. 

The  advantages  to  be  gained  from  providing  lower  ceiling  heights 
than  have  heretofore  been  found  in  general  practice  are:  (1)  a  lower 
initial  time  of  reverberation,  resulting  from  a  smaller  volume  per 
seat,  (2)  reduced  surface  areas  to  be  treated  acoustically,  permitting 
more  efficient  control  of  sound  by  shaping  the  interior  surfaces,  and 
(3)  economies  realized  in  construction  costs  through  the  elimination, 
or  reduction  in  quantity,  of  acoustic  materials  usually  required,  as 
well  as  through  the  use  of  smaller  quantities  of  ordinary  building 
material. 

Two  additional  advantages  result  from  the  proper  control  of  ceiling 
height  and  structural  volume  that  should  not  be  overlooked.  One  is 
that  economies  in  the  size  and  capacity  of  sound-reproducing  systems 
are  frequently  made  possible  in  theater  auditoriums  having  reduced 
volumes  per  seat.  The  other  advantage  is  that  excessive  power  output 
is  not  required  to  compensate  for  high  energy  losses  frequently  caused 
by  the  use  of  acoustic  materials  on  wall  or  ceiling  surfaces. 

DETAILED  ACOUSTICAL  FORM 

The  final  phase  of  theater  planning  influencing  both  the  acoustical 
condition  and  the  architectural  treatment  is  the  detailed  shaping  or 
styling  of  the  surfaces  within  the  auditorium.  This  phase  of  planning 
is  also  very  important  because  it  functions  with  the  proper  determina- 
tion of  basic  outline  and  structural  volume  to  control  the  character 
of  sound  and  the  destination  of  sound  reflections. 

When  an  auditorium  is  to  be  used  principally  for  direct  speech  or 
musical  presentations  it  is  desirable  to  plan  and  arrange  the  interior 
surfaces  so  as  to  aid  in  reinforcing  the  sound  produced  on  the  stage. 
In  the  motion  picture  theater,  however,  where  sound  is  reproduced 
and  adequate  power  can  be  provided  electrically,  the  acoustical 
problem  is  not  one  of  designing  surfaces  to  gain  reinforcement. 
Rather,  the  interior  surfaces  of  this  auditorium  should  be  shaped  and 


Feb.,  1939] 


ACOUSTICS  AND  ARCHITECTURE 


165 


arranged  so  that  they  function  to  break  up  or  disperse  sound  energy. 
This  result  can  be  accomplished  most  successfully  in  cases  where 
favorable  basic  proportions  are  maintained  and  where  the  initial  time 
of  reverberation  is  efficiently  controlled  by  the  structural  volume  of 
the  auditorium. 

Irregularity  of  surfaces  arranged  to  break  up  or  disperse  sound 
energy  may  take  the  form  of  angular  or  sloping  sections,  nonsym- 
metrical  broken  offsets,  or  convex  projections.  The  size  of  each  sur- 
face unit,  its  position  and  arrangement  on  a  wall  or  ceiling,  and  its 


CROSS     SECTION 


LONGITUDINAL      SECTION 


FIG.  3.    Theater  form  showing  angular  surfaces  for  controlling  sound  reflection. 

degree  of  projection  from  a  horizontal  or  vertical  line  will  depend 
upon  the  requirements  for  control  of  the  destination  and  dispersion 
of  sound  reflections  in  the  individual  design.  The  surface  of  a  major 
angular  or  convex  projection  may  in  turn  be  broken  into  smaller  in- 
crements if  required  in  special  cases  for  dispersion  of  the  very  high 
frequencies. 

AN  EXAMPLE  OF  FUNCTIONAL  PLANNING 

Fig.  3  shows  the  longitudinal  and  cross-sectional  views  of  a  motion 
picture  theater  planned  to  seat  900  persons.  This  theater  was 
recently  designed  in  accordance  with  the  principles  outlined  in  this 


166  C.  C.  POTWIN  AND  B.  SCHLANGER  [J.  S.  M.  P.  E. 

paper  and  is  now  under  construction.  The  horizontal  proportions 
of  the  auditorium  are  in  the  ratio  of  1 : 1.79  and  the  structural  volume 
is  123  cu.-ft.  per  seat.  No  sound-absorbing  materials  are  used  on 
either  the  wall  or  ceiling  surfaces  of  this  auditorium. 

These  surfaces  are  of  furred  construction  and  are  finished  in  ordi- 
nary hard  plaster.  The  side  walls  are  composed  of  a  series  of  horizontal 
angular  or  sloping  sections  which  vary  not  only  in  width  but  also  in 
their  degree  of  projection  from  a  vertical  line.  The  rear  wall  area 
exposed  to  the  incidence  of  sound  is  reduced  to  a  minimum  and  a 
convex  projection  is  incorporated  in  the  design  of  the  balcony  rear 
wall.  The  ceiling  surface,  which  takes  the  form  of  sloping  planes 
joined  by  convex  sections,  is  also  designed,  as  are  the  wall  surfaces, 
to  control  the  destination  and  dispersion  of  sound  reflections. 

CONCLUSION 

The  authors  have  attempted  to  show  in  this  paper  that  through  the 
proper  coordination  of  auditory,  visual,  and  esthetic  requirements, 
it  is  possible  to  plan  more  efficient  and  more  economical  theater 
structures  than  have  in  most  cases  been  designed  in  the  past. 

In  viewing  the  various  outline  forms  presented  in  the  foregoing 
study  of  typical  designs,  much  is  to  be  said  in  favor  of  plans  having 
an  upper  level  of  seating.  Such  a  plan  usually  offers  the  most 
efficient  solution  to  the  control  of  horizontal  proportions,  ceiling 
height,  and  volume  per  seat,  assuming  that  the  requirements  for 
correct  vision  are  properly  incorporated  in  fundamental  planning. 
This  form  of  design  also  introduces  a  structural  break-up  at  the 
rear  of  the  auditorium  that  is  initially  helpful  in  controlling  sound 
reflections. 

The  planning  of  detailed  acoustical  forms  for  the  wall  and  ceiling 
surfaces  offers  unusual  possibilities  for  the  creation  of  new  modes  or 
styles  in  the  esthetic  treatment  of  the  theater  auditorium.6'7  Un- 
questionably, many  highly  interesting  and  altogether  unique  designs 
will  develop  in  future  planning  when  the  functions  of  the  motion 
picture  theater  are  in  reality  adopted  as  the  inspiration  for  creative 
and  efficient  architecture. 

REFERENCES 

1  Report  of  the  Projection  Practice  Committee,  /.  Soc.  Mot.  Pict.  Eng.,  XXX 
(June,  1938),  p.  636. 

2  POTWIN,  C.  C.:    "Theater  Acoustics,"  Architectural  Record  (Building  Types 
Section)  (July,  1938),  p.  119. 


Feb.,  1939]  ACOUSTICS  AND   ARCHITECTURE  167 

3  POTWIN,  C.  C.:    "Theater  Acoustics  Today,  "Better  Theaters  (Aug.,   1937), 
p.  36. 

4  SCHLANGER,   B.:       "Motion  Picture  Theater  Shape  and   Effective  Visual 
Reception,"  J.  Soc.  Mot.  Pict.  Eng.,  XXVI  (Feb.,  1936),  p.  128. 

5  SCHLANGER,  B.:    "Cinemas,"  Architectural  Record  (Building  Types  Section), 
(July,  1938),  p.  113. 

6  "Acoustical  Forms   as    Decoration,"   Architectural  Rev.  (London),  LXXXIII 
(April,  1938),  p.  207. 

7  BAGENAL,  H.,  AND  WOOD,  A.:  "Planning  for  Good  Acoustics,"  E.  P.  Dutton 
fir  Co.  (New  York,  N.  Y.).  ( Vide  Chapt.  3,  Sec.  15,  p.  82.) 

DISCUSSION 

MR.  CRABTREE:  This  has  been  an  ideal  demonstration  of  a  collaboration 
paper,  and  likewise,  a  demonstration  of  collaboration  in  presentation.  I  have 
long  had  in  mind  the  idea  that  the  Society  should  make  definite  recommenda- 
tions for  architects  in  regard  to  (1}  architectural,  (2)  acoustical,  and  (3)  optical 
features  of  motion  picture  theaters.  It  would  seem  to  be  in  order  to  suggest 
that  our  President  appoint  a  Theater  Construction  Committee,  which  could 
draw  up  definite  recommendations.  We  already  have  the  assurance  of  the 
architectural  societies  that  after  such  recommendations  have  been  fully  discussed 
by  our  own  organization  they  would  then  publish  them  and  discuss  them  in 
their  organizations,  after  which  modifications  could  be  made  and  the  recom- 
mendations finally  drawn  up  and  circulated. 

Where  is  this  theater  being  constructed?  If  I  am  in  that  vicinity  I  will 
certainly  make  a  point  of  going  to  see  it. 

MR.  SCHLANGER:    At  Hamden,  Conn. 

MR.  CRABTREE  :  Are  there  any  in  the  vicinity  of  New  York  City  that  in  any 
way  approach  or  simulate  this  one? 

MR.  SCHLANGER:  A  theater  recently  completed  in  New  York  and  similar  in 
appearance  to  the  small  theater  on  the  French  Liner  Normandie  has  a  minimum 
auditorium  height  and  a  floor  shape  designed  in  accordance  with  the  latest 
principles. 

MR.  POTWIN:  In  duplicating  the  architectural  treatment  of  the  original 
Normandie  theater,  the  limitations  placed  on  form  made  the  acoustical  design 
somewhat  less  flexible  than  the  Hamden  project.  Nevertheless,  it  was  possible 
to  establish  a  favorable  relationship  between  the  structural  volume  and  total 
seating  of  the  auditorium  and  to  shape  and  arrange  wall  and  ceiling  splays  to 
promote  the  efficient  control  of  sound  reflections.  On  this  basis  a  minimum 
amount  of  acoustical  material  was  used,  this  material  being  confined  to  a  very 
limited  portion  of  the  rear  wall  area. 

MR.  CRABTREE  :  In  some  theaters  it  is  the  practice  to  have  intermissions  and 
throw  on  the  lights  so  one  can  see  the  beauty  of  the  decorations;  when  the  show 
is  over  they  put  on  the  lights  again.  In  this  style  of  theater  it  would  seem  almost 
imperative  that  the  lights  never  be  thrown  on,  unless  a  decorative  system  of 
color  or  spotlighting  is  provided.  In  other  words,  there  is  no  question  that  if 
the  white  lights  are  thrown  on,  such  a  theater  will  appear  somewhat  like  a  barn. 

MR.  SCHLANGER:  The  simple  architectural  forms  proposed  would  lend  them- 
selves effectively  to  color  lighting. 


168  C.  C.  POTWIN  AND  B.  SCHLANGER 

MR.  GREENE  :  Would  it  not  be  possible  to  use  a  very  simple  projected  design 
upon  the  side  walls  between  pictures,  particularly  if  you  are  not  going  to  raise 
the  level  of  illumination  high? 

MR.  SCHLANGER:  That  is  an  excellent  idea  and  has  been  done  successfully. 
Certain  types  of  cut  outs  can  be  made  and  placed  in  front  of  the  light-sources. 
Variations  in  the  images  would  make  it  possible  to  create  more  designs  per  year 
than  could  be  achieved  with  any  fixed  forms  of  decoration.  It  can  be  done 
from  the  fascia  of  the  mezzanine,  from  the  projection  room,  or  from  hidden  spots. 

MR.  CRABTREE:    Perhaps  kaleidoscopic  changes  could  be  used. 

MR.  SCHLANGER:    Yes,  that  has  been  thought  of. 

MR.  CRABTREE  :  With  regard  to  the  aisle  seats,  when  the  seats  are  staggered, 
why  is  not  the  end  seat  in  every  alternate  row  made  a  little  wider  so  the  edges 
of  the  aisles  will  be  straight? 

MR.  SCHLANGER:  If  the  seats  are  staggered  there  is  a  10-inch  difference  be- 
tween the  seat  and  the  aisle  line. 

MR.  CRABTREE:    Why  not  make  the  end  seat  about  ten  inches  wider? 

MR.  SCHLANGER:  It  would  be  out  of  proportion.  However,  we  could  com- 
promise by  making  the  seat  three  or  four  inches  wider,  so  that  the  indentation 
would  not  be  so  obvious  and  the  width  of  its  arm  block  could  also  be  increased. 

MR.  CRABTREE:  Some  theaters  furnish  seats  for  the  hard-of -hearing.  Why 
not  reserve  those  seats  for  the  corpulent? 

MR.  SCHLANGER:  No  doubt  those  seats  would  be  more  comfortable  for  the 
extremely  corpulent  persons. 

MR.  CRABTREE:  Do  you  think,  Mr.  Schlanger,  that  the  data  are  sufficiently 
far  along  that  the  Society  would  be  in  position  to  make  definite  recommendations 
if  we  had  a  committee  as  suggested? 

MR.  SCHLANGER:  The  committee  could  be  formed,  but  it  would  have  to  work 
in  close  collaboration  with  the  Projection  Practice  Committee  and  the  Sound 
Committee. 

MR.  CRABTREE:  The  committees  have  been  assembling  data  for  some  time 
past.  Do  we  have  enough  data  now,  or  do  we  have  to  do  more  experimental 
work? 

MR.  SCHLANGER:  Naturally,  more  research  work  is  always  in  order.  How- 
ever, the  new  committee  could  compile  its  own  data  and  report  to  the  other 
committees  as  to  what  additional  data  are  required  from  them  in  order  to  arrive 
at  the  proper  conclusions. 

MR.  POTWIN:  Unquestionably  with  the  data  now  available  it  should  be 
possible  to  arrive  at  definite  standards  for  a  number  of  phases  of  theater  design 
and  construction. 

MR.  CRABTREE:  If  the  committee  did  nothing  more  than  prevent  the  archi- 
tects from  placing  lights  right  under  one's  nose,  I  think  the  effort  would  be  well 
worth  while. 


[ARACTERISTICS  OF  FILM  REPRODUCER  SYSTEMS* 

F.  DURST  AND  E.  J.  SHORTT** 

Summary. — An  analysis  of  sound-picture  reproducing-system  characteristics,  in- 
cluding electrical  and  acoustical  response  data  collected  in  the  interest  of  determining 
the  possibilities  involved  in  obtaining  an  average  characteristic  for  reproducing  vari- 
ous film  products  with  uniform  response  over  several  combinations  of  loud  speaker 
equipment.  With  the  aid  of  a  curve  tracer  having  a  long-persistent  cathode-ray 
screen,  a  photographic  record  was  made  of  the  characteristics,  starting  with  various 
forms  and  amounts  of  equalization  and  exploring  their  relationship  to  the  power- 
handling  capacity  of  amplifiers.  Following  through  the  system,  this  record  shows  the 
characteristics  of  dividing  networks  under  various  conditions  of  load,  and  finally  the 
acoustical  response  curves  taken  for  comparison  of  the  loud  speaker  equipments  under 
study. 

The  measurements  of  loud  speaker  combinations  included  various  types  of  units, 
both  permanent-magnet  and  energized,  low-frequency  horns  ranging  from  open  back 
baffles  to  folded  horns  with  specially  designed  rear-loading  compartmsnt,  and  high- 
'  frequency  multicellular  horns  of  various  configurations  and  constructional  details. 

After  establishing  the  natural  characteristics  of  the  various  equipments  involved, 
careful  listening  tests  were  made  over  an  extended  period  with  samples  of  commercial 
prints  and  other  recordings.  A  description  follows  of  the  difficulties  and  problems 
involved  in  an  effort  to  obtain  one  overall  chiracteristic,  which  would  give  satisfactory 
reproduction  for  all  types  of  material.  The  final  results  are  shown,  with  a  short  dis- 
cussion of  the  methods  for  duplication  in  other  equipment  combinations,  and  con~ 
elude  with  recommendations  for  future  designs  and  ratings. 

Our  interpretation  of  the  goal  of  all  organizations  concerned  with 
sound  in  the  motion  picture  industry  is  to  produce  in  every  theater 
throughout  the  world  the  sounds  considered  desirable  by  the  director 
or  whoever  is  responsible  for  the  production.  In  order  to  achieve 
this  goal,  it  is  obvious  that  the  performance  of  the  equipment  must  be 
so  well  understood,  and  so  well  standardized,  that  the  director  in 
his  review  room  can  hear  his  product  as  his  customers  will  hear  it. 

It  would  be  very  convenient  if  we  could  standardize  the  perform- 
ance of  theater  equipment  by  specifying  that  it  should  have  a  flat 

*Presented  at  the  1938  Fall  Meeting  at  Detroit,  Mich.;   received  November 
18,  1938. 

*  International  Projector  Corp.,  New  York,  N.  Y. 

169 


170 


F.  DURST  AND  E.  J.  SHORTT 


[J.  S.  M.  p.  E. 


w> 
I 

'-5 


Feb.,  1939]  FILM  REPRODUCER  SYSTEMS  171 

gain-frequency  characteristic ;  that  it  should  introduce  no  non-linear 
distortion  at  any  and  all  levels;  and  that  the  distribution  should  be 
uniform  to  every  seat  in  the  theater.  To  adhere  even  approximately 
to  such  a  standard  would,  however,  work  a  hardship  on  both  the 
theater  owner  and  the  producer.  It  would  make  the  theater  owner 
pay  more  than  is  necessary  or  desirable  for  him  to  pay  for  his  equip- 
ment, and  the  best  product  that  the  producer  could  turn  out  would 
suffer  in  signal-to-noise  ratio  and  the  general  acceptability  of  the 
quality. 

By  using  film  as  a  recording  method,  we  accept  a  medium  in  which 
most  of  the  annoyance  from  noise  is  concentrated  in  the  high-fre- 
quency end  of  the  spectrum.  Most  of  the  unpleasant  distortion 
experienced  in  film  reproduction  also  occurs  at  the  same  end  of  the 
frequency  characteristic.  Overall  performance  can,  therefore,  be 
improved  by  utilizing  a  reproducing  system  characteristic  in  which 
the  response  falls  off  at  the  high  frequencies.  If  the  producer  is  to 
use  noise-reduction  of  the  type  most  generally  available,  the  low- 
frequency  response  of  the  reproducing  system  should  be  limited  so 
that  the  listener  will  not  be  annoyed  by  hearing  the  operation  of  the 
noise-reduction  device.  With  this  objective  in  mind,  it  appears  that 
the  reproducing  system  characteristic  should  be  standardized  in  such 
a  way  as  to  derive  the  maximum  benefit  in  signal-to-noise  ratio  and  in 
the  reduction  of  undesirable  distortion,  considering  at  the  same  time 
the  theater  owner's  interest  in  reduced  cost  of  equipment.  The 
Research  Council  of  the  Academy  of  Motion  Picture  Arts  and 
Sciences,  through  the  work  of  its  Committee  on  Standardization,  has 
established  a  basis  for  the  desired  attainment.  It  has  brought  out 
the  desirability  of  certain  characteristics  that  must  be  carefully  con- 
sidered from  every  angle.  In  certain  instances,  some  of  these  charac- 
teristics might  be  misinterpreted  unless  proper  consideration  is 
given  to  the  variable  factors  involved. 

In  Fig.  1  an  effort  has  been  made  to  indicate  diagrammatically  the 
various  steps  of  the  overall  recording  and  reproducing  process.  It 
will  be  noted  that  variations  exist  throughout  the  entire  chain.  In 
the  recording,  consideration  must  be  given  to  the  acoustics  of  every 
set.  Microphone  placement  has  not  only  been  a  point  of  con- 
troversy, but  has  always  been  a  factor  of  adjustment  in  the  overall 
characteristic.  The  technic  of  mixing  is  another  vital  point  which  is 
entirely  an  adjustable  feature.  Determination  of  the  proper  amount 
of  equalization  and  the  general  amplifier  characteristics  are  additional 


172 


F.  DURST  AND  E.  J.  SHORTT 


[j.  s.  M.  p.  E. 


CLCMCNTS                                      SLOCIT  XHCMATIC 

Mxi&Lirr  HtouiiHD  m  OSTAIN  oes/peo  aespOHSf 

or^mma^ 

STUDIO 
aecoaoiNe 

ACOUSTICS  OF  SfT 
M/CROPHONCS                                                   O      O      O 
VOiUMf  CONTROLS                                             f     t'    f 

SfT  CONSraiKTION:  IHOOOfS  -( 
PLAClMtHT:  TYPtS 
tALANCt  OF  HUXIN6 

WTiMMI 

VStD 
DIALOGUC               Ix^ 

paesiNce          1          s~^ 

n>c  touAuze*              <^ 

HI6H  fASS              \j 

H.F  CO*P.eCTION  [              <^ 

L.F  CORACCTION       ^^^' 

pofT-iauju.ite*  1           ., 

J 

1 

SCOOPING                                                         M 

fCAMNING                                                  0 
SOUND  fffCOffO                                 SiX^ 
PHOTO  CfLL    COUPLING                       C/ 

SLIT  SIZl:  ILLUMINATION 
SWGLf  TRACK  ' 

PUSH  -PULL;  too  -zoo  MIL 

IMPlDAHCt  MATCHING 

AMPLIF/CP 

fti  rca                                      **         |         | 

GAIN:  POWIK;  fto/sf  fffoacrttw 

MO  Oil  I  A  TOO                                             ^ 

c 

* 

VAfflAOLf  OCNSITY:  VAtt/A&if  AfffA 

^ 

"1  

LABOBATORY 
paocesstNC                                      /!V 

PBIHT/HG                                                                  /    \ 

J 

THfJTff? 

SLIT  SITS'  ILLUMINATION                                                              / 

• 

ffPftODUCTION                                                              •§ 
SCANMHG                                                                     ^               —  | 

PHOTO  CfLL   COUPLING                                          ^^ 

IMPlDANCe  MATCHING 
GAIN  '  POWCS;  NOIi{  aiDUCTION 

CHAXACTtBlfTK         v 

l5TOJ~ 

LOUD  spgA/re/fs                     \  JT~4  1  i~~l  1  1  

ACOUSTICS  OF  AUDITOKIUM 

-  IMPOSfO  LIMITATION-  LOUO  iPftKf*  CMBMTtRlSTK     \,  ^ 
THCtTKf  CONSTRUCTION                                                    ^  

0 

NO  KS:Z~-TS 

*    MOTf:-  OfFfaeNCf,    F.  t.  HOtPfff:    'flftr&CAl  ttfriVOaKS  FOP  SOUNO  BfCO0DING,  '  J  S  M  P.f.  /I/Of.  1936 


Ficf  2.     Flexibility  and  limitations  in  recording  and  reproducing. 


Feb.,  1939]  FILM  REPRODUCER  SYSTEMS  173 

tools  by  which  the  recording  personnel  may  adjust  the  final  film 
characteristic  to  afford  the  most  pleasing  and  dramatic  reproduction 
in  the  theater. 

In  the  laboratory  is  found  a  situation  that  contributes  further  to 
the  variations  in  film  characteristics.  It  is  understood  that  for 
extreme  conditions  high-frequency  losses  may  range  from  0  to  8  db. 
at  8000  cycles.  While  a  considerable  proportion  of  this  may  be 
anticipated  in  the  adjustment  of  recording  characteristics,  the  power 
ratios  for  unanticipated  variations  are  important,  as  they  occur  in 
the  high-frequency  range.  Minimizing  printer  slippage  is  a  very 
pertinent  factor  in  any  attempt  to  obtain  clean  and  smooth  high-end 
reproduction  free  from  harshness. 

The  principal  variables  in  reproducing  systems  may  be  grouped 
under  the  headings  of  flutter,  electrical  characteristics,  loud  speaker 
characteristics,  and  theater  acoustical  conditions.  While  the 
electrical  characteristics  have  heretofore  been  established  as  extend- 
ing from  film  input,  including  the  optical  system,  to  a  resistance 
termination  at  the  output  of  the  amplifier,  it  is  proposed  to  show 
here  the  inadequacy  of  such  standards.  Even  measurements  to  voice- 
coil  terminations  are  not  a  true  indication  of  the  influence  of  the  horn 
equipment  and  the  acoustical  coupling  on  the  aural  response  that  is 
sold  to  our  customers. 

In  our  interest  of  standardization  in  the  theater,  extensive  tests 
were  made  with  various  types  of  reproducing  equipment.  If  systems 
generally  are  manufactured  to  conform  to  specified  requirements,  it 
must  follow  that  similar  results  should  be  expected  in  the  same 
acoustical  environments.  The  differences  actually  encountered,  and 
the  reasons  therefor,  were  indicative  of  the  present  necessity  for  most 
careful  consideration  of  the  entire  program  of  standardization. 
While  it  was  disturbing  to  note  the  large  amount  of  variation  existing 
in  the  characteristics  of  film  products  today,  it  is  believed  that  the 
hope  of  remedying  this  situation  lies  in  the  establishment  of  specific 
end  results  to  which  recording,  laboratory,  and  reproducing  groups 
may  work. 

Fig.  2  shows  a  tabulation  of  an  outline  wherein  desirable  elements 
of  flexibility  are  maintained  for  recording,  in  direct  contrast  to  the 
present  imposed  limitations  in  the  reproducing  group.1  In  order 
to  establish  a  coordinated  program  for  overall  betterment,  each 
group  must  assume  its  rightful  share  of  responsibility,  and  be  per- 
mitted complete  control  over  matters  within  its  own  jurisdiction. 


174 


F.  DURST  AND  E.  J.  SHORTT 


[J.  S.  M.  P.  E. 


In  the  laboratory,  it  is  suggested  that  further  studies  be  made,  par- 
ticularly in  the  interest  of  reducing  the  ill  effects  of  printer  slippage 
and  of  maintaining  closer  control  of  high-frequency  losses.  In  the 
theater  field,  it  is  desirable,  in  the  interest  of  our  customers,  to  be 


S      30  fOO  300  fKc  fOKC 

FREQUENCY  IN  CYCLES  P£R  SfCONO 

FIG.  3.     Overall  acoustic  characteristic.     (Adjusted  for 
most  pleasing  response.) 


allowed  freedom  from  component  or  sectionalized  specifications. 
End  results,  not  details,  are  the  logical  basis  for  performance 
specifications. 

As  a  general  arrangement,  our  own  tests  indicated  that  a  flat 
overall  acoustic   characteristic   was   not  always   the  best.     Fig.    3 


'LATIVC  SOUND  INTCNSITY  IN  DB 
\  \  + 
5  *»  o  o, 

^  „ 

A, 

OU 

ecr 

w 
we 

CA 
AL 

''$ 

rm**^ 

/ 

X. 

•\ 

tL 

WORK-/ 

s 

\ 

tOO  300  /AC 

FR£QU£NCY  IN  CYCLfS  f>CR  SECONDS 


FIG.  4.    Overall  acoustic  characteristic.    (Showing  acous- 
tic and  electrical  modifications.) 


shows  the  acoustic  characteristic  found  to  be  most  desirable  in  this 
particular  theater. 

When  comparison  was  made  with  similar  curves,  obtained  from  a 
number  of  widely  scattered  theaters  wherein  the  quality  of  reproduc- 
tion has  been  approved  by  various  technical  groups,  it  was  noted  that 


Feb.,  1939]  FILM  REPRODUCER  SYSTEMS  175 

a  flat  overall  acoustic  response  did  not  qualify  as  giving  the  most 
pleasing  reproduction.  It  was  observed  in  general  that  the  level 
in  the  region  of  250  cycles  was  approximately  6  db.  lower  than  the 
balance  of  the  characteristics. 

Upon  analysis  it  was  found  that  such  a  characteristic  as  shown  in 
Fig.  2  creates  an  impression  of  roundness  or  fullness  of  the  bass  notes 
without  causing  too  much  heaviness  in  the  region  where  chest  tones 
become  objectionable.  From  this  the  natural  deductions  would 
point  to  insufficient  dialog  equalization  in  recording,  but  such  con- 
clusions can  not  be  established  until  the  studio  review  room  acoustic 
characteristics  are  known.  Our  analysis  was  made  by  direct  A-B 
comparison  with  a  system  whose  overall  acoustical  response  was 
intentionally  made  flat  through  this  region,  as  shown  in  Fig.  4. 
This  figure  shows  also  the  effect  of  acoustic  and  electrical  networks 
that  may  be  used  to  alter  the  characteristic.  In  passing,  it  may  be 
pointed  out  that  the  electrical  network  offers  considerable  advantage 
in  that  it  can  be  made  adjustable  at  very  low  cost. 

It  is  interesting  to  observe  how  this  "sway-back"  characteristic 
has  been  obtained  by  using  component  parts  which  in  themselves  all 
have  straight  lines  in  the  "sway-back"  region.  A  study  of  these 
conditions  was  made,  using  the  RCA  curve  tracer2  provided  through 
the  courtesy  of  Mr.  L.  C.  Hollands  of  the  Radiotron  Division  of  RCA 
in  Harrison,  N.  J.  Figs.  5,  6,  and  7  show  the  measured  acoustical 
characteristic  of  a  system  wherein  the  straight-line  elements  were 
connected  together,  each  in  a  different  manner.  In  each  case  a 
dividing  network,3  designed  to  operate  between  equal  impedances, 
was  used,  and  the  high-frequency  branch  terminated  by  a  load  of  the 
proper  value.  The  measured  electrical  characteristics  of  this  net- 
work have  been  plotted  on  each  curve. 

In  Fig.  5  the  low-frequency  units  were  arranged  to  provide  a 
matched  impedance  termination.  In  Fig.  6  the  low-frequency  load 
was  changed  so  that  the  impedance  ratio  between  the  network  and  the 
load  was  12  to  6.  In  Fig.  7  the  low-frequency  units  were  arranged 
so  that  the  impedance  ratio  was  12  to  1.5.  These  figures  show  the 
effect  of  impedance  mis-matching  of  the  network  and  the  loud 
speakers,  and  indicate  the  importance  of  impedance  relationships 
and  the  definite  necessity  of  providing  standards  that  are  more  in- 
clusive than  the  present  form  of  electrical  characteristics  which  are 
measured  across  resistance  terminations  at  the  amplifier  output. 
The  same  overall  response  can  be  obtained  by  redesigns  of  the  net- 


176 


F.  DURST  AND  E.  J.  SHORTT 


[J.  S.  M.  P.  E. 


60  200  400        IKC 

ftfOUCNCY/N  CYClfS  PCR  SfCOND 


IOKC 


FIG.    5.     Overall    acoustic    characteristic.        (Network 
output  12  ohms,  l.-f.  load  12  ohms.) 


o  eo         200       400     me  tone 

FZ£QU£NCY  IN  CYCLES  Pfff  5£CONO 

FIG.    6.     Overall    acoustic    characteristic.        (Network 
output  12  ohms,  l.-f.  load  6  ohms.) 


^oo        400     we 

FttQUCNCY  IN  CYSL£5  f>£ff  SfCOND 


tOKC 


FIG.     7-     Overall     acoustic     characteristic.     (Network 
output  12  ohms,  l.-f.  load  1.5  ohms.) 


Feb.,  1939]  FlLM  REPRODUCER  SYSTEMS  177 

work  wherein  the  low-frequency  branch  is  purposely  made  to  have  a 
higher  impedance  than  the  load  into  which  it  will  work,  or  by  reduc- 
ing the  voice-coil  impedance  of  the  low-frequency  units. 

In  reviewing  numerous  acoustical  response  characteristics,  varia- 
tions in  the  upper  end  of  the  frequency  spectrum  were  noted.  The 
degree  of  tolerance  for  highs,  the  acoustical  condition  of  the  theater, 
the  efficiency  and  response  characteristics  of  the  horns  and  loud 
speaker  units,  the  effect  of  printer  slippage,  flutter,  transient  distor- 
tion resulting  from  too  sharp  cut-off  niters,  resonance  conditions,  or 
phase  displacements  may  all  have  been  contributing  factors.  Our 
tests  disclosed  positive  differences  in  the  response  characteristics  of 
various  types  of  horn  equipment. 

A  substantially  flat  system4  is  probably  the  most  desirable  type 
to  use  in  the  theater  for  reasons  of  economy  of  construction  and 
simplification  of  general  maintenance  and  tuning-up  procedures. 
It  is  believed  that  specifications  should  not  require  complicated, 
costly,  or  critical  designs  in  theater  systems,  particularly  in  view  of 
the  flexibility  at  present  available  in  recording  equipment.  Further- 
more, the  cost  of  modifying  recording  systems  to  insert  therein  any 
desired  characteristics  and  so  maintain  a  desirable  response  in  the 
theater  would  appear  to  be  more  economical  for  the  industry  as  a 
whole  due  to  the  relative  number  of  equipments  involved. 

Mention  has  previously  been  made  of  the  necessity  for  maintain- 
ing the  desired  characteristic  in  the  theater  at  any  and  all  sound 
levels.  It  will  be  readily  appreciated  that  this  problem  lies  beyond 
the  scope  of  this  paper  as  it  must  include  data  on  various  types  of 
amplifier  design,  involving  triodes,  beam  power  tubes,  pentodes,  with 
and  without  feed-back,  and  various  types  of  feed-back  circuits. 

With  regard  to  the  relationship  between  power  output  and  fre- 
quency response,  a  composite  picture  (Fig.  8)  was  made  to  show  how 
a  given  amplifier  characteristic  was  altered  when  that  amplifier  was 
pushed  to  its  maximum  output.  In  this  particular  instance,  both 
ends  of  the  frequency  spectrum  were  initially  raised  about  6  db. 
above  the  1000-cycle  level.  It  will  be  noted  that  as  the  output  of 
this  amplifier  was  increased,  the  characteristic  approached  a  "ceiling" 
and  finally  flattened  out. 

Fig.  9  shows  the  overdrive  characteristic  of  the  same  amplifiei 
for  various  frequencies.  It  will  be  observed  that  in  this  particular 
case  the  various  curves  show  very  little  departure  from  each  other 
Overdrive  characteristics  of  another  amplifier  are  shown  in  Fig.  10. 


178 


F.  DURST  AND  E.  J.  SHORTT 


[J.  S.  M.  P.  E. 


JTPUT  LfVCL  IN  DB 

*«*_* 

0 

Q 

* 

rf  1? 

—  . 

"" 

k^\ 

h-. 

r^ 

* 
/ 

,  — 

ss 

5   *» 

^x^"1*-^ 

-^ 

^/ 

O    $ 

K>0  300  IKC 

FRfQUCNCY  IN  CYCifS  P£R  SfCOND 


FIG.    8.     Frequency   response   characteristic.     (For   various 
output  levels.) 


1,000 
5,000 

Z  00 
0,000 


O    /     2     3     4     S     67     S     9     to    ft    Ig    13 

INPUT  INCR£AS£  IN  DQ 

FIG.  9.  Overdrive  power-frequency  character- 
istics. (Feed-back  type.)  Abscissas  represent 
input  increase  in  db.  above  rated  load. 


OUTPUT  LfVfL  IN  OB 

<OO-»l\)lM4tlxOi  vi 

\ 

^x- 

' 

0,000 

S.QOO 

^ 

^ 

•^ 

^. 

^ 

+*£- 

/ 

s" 

^ 

r^- 

I0i 

to 

/£ 

<£~ 

•**" 

fo< 

•> 

~? 

^» 

/ 

•)   i    2   j   4    s   e   7    a  9  to  ft  tz  a 

INPUT  lfJCff£ASf  /N  DB 

FIG.  10.  Overdrive  power-frequency  charac- 
teristics. (Other  types  of  theater  amplifiers.) 
Abscissas  represent  input  increase  in  db.  above 
rated  load. 


Feb.,  1939]  FILM  REPRODUCER  SYSTEMS  179 

Here  it  will  be  noted  that  as  the  input  level  is  increased  the  low- 
frequency  power  output  will  not  increase  beyond  its  compression 
point,  but  at  the  higher  frequencies  it  will  continue  to  deliver  more 
and  more  power.  Under  conditions  of  extreme  overdrive,  it  may  be 
expected  that  the  power  delivered  by  such  an  amplifier  to  the  high- 
frequency  units  may  be  two  to  three  times  the  amount  that  can  be 
delivered  to  the  low-frequency  units.  With  this  situation  it  would 
be  particularly  difficult  to  maintain  any  predetermined  character- 
istic in  a  theater  at  all  levels.  With  demands  for  increased  power 
output  to  be  provided  in  order  to  obtain  the  proper  dramatic  effect 
for  various  types  of  recordings,  the  tendency  to  overdrive  amplifiers 
will  inevitably  grow.  It  appears,  therefore,  that  standards  for  elec- 
trical power  must  give  these  factors  further  consideration. 

With  respect  to  power  ratings  and  requirements,5  attention  is 
called  to  the  block  diagram  of  power  amplifiers  shown  in  Fig.  1. 
In  the  upper  grouping,  the  amplifiers  are  paralleled  ahead  of  the  divid- 
ing network.  The  total  wattage  available  would  naturally  be  the 
algebraic  sum  of  the  individual  amplifiers,  and  if  no  attenuation  is 
provided  in  the  dividing  network  both  high-  and  low-frequency 
units  will  obtain  that  same  total  amount  of  power  in  their  respective 
frequency  ranges.  In  the  lower  group,  where  the  dividing  network  is 
inserted  ahead  of  the  power  amplifiers,  it  is  common  practice  to  use 
less  power  in  the  high-frequency  branch  by  reason  of  the  relative 
efficiencies  of  the  high-  and  low-frequency  loud  speaker  units.  It 
has  come  to  our  attention  that  in  discussions  of  power  ratings  for 
volume  or  seating  requirements  the  total  power  in  these  two  branches 
has  been  used,  while  in  reality  the  maximum  power  does  not  exceed 
the  amount  that  is  diverted  in  the  low-frequency  branch  alone. 
This  illustration  is  made  as  further  proof  of  the  possibilities  of  mis- 
interpretation of  standards,  and  at  the  same  time  to  indicate  addi- 
tional reasons  for  adopting  acoustical  standards  throughout. 
Without  a  doubt  the  feeling  exists  that  acoustical  measurements 
e  meaningless,  and  that  adequate  test  equipment  is  not  now 
ailable.  Comparison  of  our  own  work  with  the  results  obtained 
y  others  prevent  our  subscription  to  this  theory  in  its  entirety 
Improvements  can  and  always  will  be  made,  but  our  work  in  this 
field  has  led  us  to  believe  that  the  present  facilities  provide  a  far  more 
accurate  indication  of  comparative  response  characteristics  than  any 
electrical  measurements,  particularly  those  of  the  spot-frequency 
variety. 


an 
av 
by 

-w- 


180  F.  DURST  AND  E.  J.  SHORTT  [j.  s.  M.  p.  E. 

In  conclusion,  it  is  evident  that  if  our  goal  of  standardized  repro- 
duction is  to  be  achieved,  investigations  should  be  continued  con- 
cerning the  overall  characteristics  of  reproducing  systems.  It  is 
firmly  believed  that  to  carry  this  out  completely,  and  to  avoid  the 
pitfalls  of  misinterpreting  component  part  specifications,  overall 
acoustical  performance  characteristics  must  be  the  criteria,  and  final 
acceptance  tests  and  tune-up  procedure  must  be  based  upon  the  use 
of  the  warble  film  and  acoustical  measurements,  rather  than  sec- 
tionalized  response  tests.  It  is  believed,  in  this  connection,  that 
the  manufacturer  should  be  allowed  to  decide  the  methods  by  which 
he  will  provide  this  final  characteristic.  Basically,  the  problem 
resolves  into  determining  the  best  overall  acoustical  characteristic 
to  which  film  producers  can  most  readily  and  consistently  adjust 
their  product,  at  the  same  time  keeping  the  cost  to  exhibitors  at  a 
minimum.  Last  but  not  least,  it  is  essential  that  the  listening 
facilities  of  both  patrons  in  the  theaters  and  the  producers  in  their 
review  rooms  be  made  equivalent  in  order  that  they  may  obtain  the 
same  reactions. 

Appreciation  for  his  cooperation  is  expressed  to  Mr.  J.  B.  Sherman 
of  the  RCA  Radio tron  Division,  Harrison,  N.  J. 

REFERENCES 

1  HOPPER,  F.  L.:     "Electrical  Networks  for  Sound  Recording,"  /.  Soc.  Mot. 
Pict.  Eng.,  XXXI  (Nov.,  1938),  p.  443. 

2  SHERMAN,  J.   B.:    "An  Audio-Frequency  Curve  Tracer,"  Proc.  I.R.E.,  26 
(June,  1938),  No.  6,  p.  700 

3  "Dividing  Networks  for  Loud  Speakers,"  Technical  Bulletin  (March  3,  1936), 
Academy  of  Motion  Picture  Arts  &  Sciences,  Hollywood,  Calif.      (cf.  Fig.  d,  p. 
23). 

4  "Standard  Electrical  Characteristics,"  Technical  Bulletin   (June  8,   1937), 
Academy  of  Motion  Picture  Arts  &  Sciences,  Hollywood,  Calif,    (cf.  p.  3). 

6  "Procedure  for  Projecting  'Hi-Range'  Prints  in  the  Theater,"  Technical 
Bulletin  (Nov.  24,  1937),  Academy  of  Motion  Picture  Arts  &  Sciences,  Hollywood, 
Calif,  (cf.  p.  5). 

DISCUSSION 

MR.  GOLDSMITH  :  It  is  very  obvious  from  these  interesting  curves  which  have 
been  presented  with  unusual  completeness  that  engineering  compromises  must 
be  made  in  reproducing-equipment  design  in  order  to  give  reasonably  uniform 
results  in  the  theater  for  all  types  of  product  and  on  an  economic  basis.  If  the 
engineer  were  to  specify  "perfect"  acoustics  from  zero  cycles  to  the  highest  fre- 
quency which  we  hear,  it  would  require  that  the  theater  be  correct  architecturally 
and  acoustically,  that  the  film  be  kept  in  immaculate  condition  to  eliminate  all 


Feb.,  1939] 


FILM  REPRODUCER  SYSTEMS 


181 


film  noises,  and  that  other  ideal  conditions  must  hold.  Inasmuch  as  the  condi- 
tions which  are  encountered  in  actual  practice  can  not  approach  the  ideal,  broad 
compromises  must  be  made.  It  would  be  interesting  to  know  whether  these 
performance  curves  have  been  selected  as  a  matter  of  engineering  judgment,  in 


i 

3; 


Xs 


30 


K>0 


300  //rc 

FREQUENCY  /V  CYCLES  PER  SECOUD 


tOKC 


FIG.  11.     Overall  system  acoustic  characteristic. 

Lview  of  economic  conditions  and  considering  what  studios,  theaters,  and  a  vail  - 
i  able  recording  and  reproducing  equipment  could  and  would  do,  and  what  quality 
|  of  sound  film  can  reproduce  without  excessive  maintenance. 

MR.  FRIEDL.     Yes,  it  is  true  that  these  curves  have  been  established  with 
that  practical  consideration.     In  continuing  this  study  of  overall  performance,  the 


V 


\ 


3O  IOO  3OO  fKC 

FREQUENCY  /N  CYCLES  PER  SECOND 

FIG.  12.     Overall  system  acoustic  characteristic. 


tone 


end  result — that  is,  bringing  to  the  patron  in  the  theater  a  better  record  or  re- 
production of  the  original — we  urge  more  sincere  cooperation  of  the  recording, 
processing,  and  reproducing-equipment  groups. 

Just  as  the  "Recommendations  on  Theater  Sound  Reproducing  Equipment" 
(prepared  by  the  Research  Council  of  the  Academy  of  Motion  Picture  Arts  and 
Sciences),  as  read  before  the  Society  in  Washington,  outlined  many  desirable  con- 
trol factors,  we  would  like  to  follow  up  and  urge  control  factors  in  other  links  of 


182  F.  DURST  AND  E.  J.  SHORTT  [j.  s.  M.  P.  E. 

the  chain.  At  the  present  time  the  manufacturer  of  the  theater  equipment 
bears  the  responsibility  of  making  acceptable  to  the  patron  the  product  delivered 
to  the  projection  room  in  the  film  can.  There  is  a  certain  latitude  and  limit  to  how 
much  the  projectionist  or  the  sound  reproducing  equipment  can  compensate  for 
recording  and  processing  variations,  particularly  inasmuch  as  economies  are  re- 
garded more  consciously  at  the  theater  end  of  the  chain  than  any  place  else  in  the 
industry.  The  composite  chart  that  you  saw  at  the  beginning  of  the  paper  con- 
tains many  more  factors  than  we  are  able  to  discuss  at  this  time. 

MR  GOLDSMITH:  Referring  to  Fig.  1,  what  is  meant  by  the  "lens  characteris- 
tic"? Do  you  mean  the  high-frequency  loss  which  results  from  the  relationship 
between  the  slit  width  and  the  length  of  the  recorded  wave  on  the  film? 

MR.  DURST:  Yes.  This  is  a  fixed  loss.  In  addition,  there  are  a  number  of 
other  variables  in  most  recording  and  reproducing  systems.  I  have  itemized  a 
number  of  them  in  Fig.  2. 

MR.  GOLDSMITH:  This  is  a  paper  of  a  helpful  type  because  it  shows  how  much 
work  remains  to  be  done  in  system  improvement,  and  frankly  gives  details. 

MR.  DURST  :  It  is  an  appeal  in  one  respect  to  the  industry  to  determine  what 
type  of  characteristic  is  best,  accept  it  as  such,  and  lay  it  down  in  such  a  fashion 
that  all  concerned  may  work  to  it  to  their  best  advantage.  It  is  very  disturbing  to 
find  the  very  great  differences  that  we  do  experience  in  the  field.  When  a  system 
is  carefully  tuned  up  for  one  picture,  the  quality  of  reproduction  alters  materially 
when  the  program  is  changed. 

MR.  ROBERTS:  I  am  interested  in  the  small  column  in  Fig.  1  entitled  "Labora- 
tory." I  would  like  a  little  more  information  on  that  one  variation  in  high  fre- 
quency. What  does  the  8-db.  loss  represent? 

MR.  DURST:  The  variations  that  may  occur  in  attention  of  the  high-frequency 
response  characteristic  range  anywhere  from  0  to  8  db. 

MR.  ROBERTS:     From  laboratory  to  laboratory? 

MR.  DURST:  Yes,  from  laboratory  to  laboratory;  or  it  is  possible  that  it  may 
vary  to  that  extent  in  any  one  laboratory. 

MR.  ROBERTS:     In  any  one  reel? 

MR.  DURST:  Not  necessarily  any  one  reel,  but  from  one  picture  to  another, 
or  from  one  product  to  another. 

MR.  WOLF:  I  was  shocked  that  there  was  such  an  overall  variation,  and  that 
the  system  had  so  much  response  in  the  low  end.  I  would  like  to  know  a  little 
more  about  the  theater  where  this  response  was  measured.  I  assume  you  con- 
sider the  response  curve  about  as  fine  a  one  as  you  know  of  in  your  practical  ex- 
perience in  the  theater.  Referring  to  the  first  curve,  Fig.  3,  the  overall  character- 
istic indicated  rather  uniform  response  except  around  the  150-  to  400-cycle  region, 
where  it  took  a  very  decided  dip  for  4  or  5  db.  and  then  went  up  again. 

That  is,  I  take  it,  an  overall  acoustic  characteristic,  measuring  the  output 
acoustically  in  the  theater.  Is  that  right? 

MR.  DURST:    Yes. 

MR.  KELLOGG  :  It  is  theoretically  impossible  to  make  up  by  choice  of  frequency 
characteristic,  for  such  factors  as  monaural  pick-up,  difference  between  micro- 
phone distance,  and  the  impression  of  distance  that  the  theater  patron  gets, 
and  the  fact  that  most  sound  is  reproduced  at  unnaturally  high  levels.  Neverthe- 
less, there  is  unquestionably  a  frequency  characteristic  which  on  the  average  gives 


Feb.,  1939] 


FILM  REPRODUCER  SYSTEMS 


183 


the  best  illusion,  or,  shall  we  say,  makes  the  unnaturalness  least  conspicuous. 
This  optimum  characteristic  no  doubt  depends  on  the  factors  just  mentioned  and 
others,  but  we  do  not  know  enough  to  figure  it  out.  We  must  determine  it  by 
trial.  It  is  simply  a  matter  of  taste. 


tOKC 


FIG.  13.     Overall  system  acoustic  characteristic. 

MR.  DURST:  That  is  very  true.  In  this  instance  a  single  microphone  was 
used  and  the  nodal  points  were  explored  to  obtain  readings  of  maximum  intensity. 
It  is,  for  purposes  of  comparison,  approximately  equivalent  to  what  might  other- 
wise have  been  obtained  with  a  warble  frequency  and  fixed  microphones.  The 
measurements  were  made  hi  the  auditorium  approximately  40  feet  from  the 
speaker  system. 


+  3 
0 

-  5 

^ 

\ 

i 

PCA  WARBLE  FREQ.  F/LM  4236-1 
G£N.RAD/O  SOUND  L£V£L  MfTCR  #7S9-A 

/ 

S 

«_ 

v_ 

/ 

\ 

7 
\ 

* 

/ 

S 

*N—  X^*^ 

N 

~~- 

X 

ELECTRICAL  CHAXACT£ftlSTK-f    7** 
^ACOUSTICAL  CH*fiACr£R/STtC  -f 

\ 

s 

\ 

\ 

\ 

1H 

30 


too 


300  fKC 

FRfQMNCY  WCYCl.£S  f>£fi  SfCOfJO 


fOKC 


FIG.  14.  Overall  system  acoustic  characteristic  (Paramount  Theater,  Los 
Angeles,  Calif.,  ERPI  Mirrophonic  system;  one  86  amplifier  and  two  87 
amplifiers).  Tests  made  by  the  Sound  Department  of  Paramount  Studio. 


MR.  WOLF:  How  did  you  integrate  the  sound  without  a  warble  tone  or 
multiple  microphones? 

MR.  DURST:     By  exploration  of  the  wave  pattern. 

MR.  WOLF:     You  picked  one  point  that  did  not  seem  to  have  any  low  points? 

MR.  DURST:  No,  the  microphone  was  moved  until  a  maximum  reading  was 
obtained  for  each  frequency  reproduced  from  a  standard  constant-frequency  film. 


184  F.  DURST  AND  E.  J.  SHORTT  [j.  s.  M.  P.  E. 

MR.  WOLF:     It  was  the  average  of  several  positions? 

MR.  DURST:  No;  it  was  the  maximum  reading  obtained  at  any  point  in  that 
vicinity.  It  was  not  a  predetermined  position,  but  the  wave  was  explored  for  the 
maximum  peak  or  signal. 

MR.  FRIEDL:  As  a  matter  of  general  interest,  Figs.  11-16*  show  measure- 
ments in  various  theaters  and  also  in  a  listening  room,  which  were  taken  with 
a  warble  film  and  which  would  give  comparable  results.  You  will  note  a  wide 
divergence  of  "end  results,"  yet  each  of  these  represents  a  listening  condition 
that  is  approved  by  a  competent  group  of  technical  people. 

MR.  WOLF:     I  never  saw  a  group  that  agreed. 

MR.  FRIEDL:  A  significant  thing  is  the  dip  hi  the  region  of  200  to  400  cycles, 
because  it  is  in  this  band  that  various  levels  of  energy  might  make  dialog  as  pres- 
ently recorded  sound  very  heavy.  Such  heaviness  of  dialog  spoils  naturalness 
and  intelligibility. 

MR.  GOLDSMITH:  A  "booming"  effect  may  result  if  the  energy  is  not  attenuated 
in  that  region. 

MR.  FRIEDL:  These  are  all  two-way  systems.  An  interesting  thing  to  note 
as  you  review  dissertations  and  presentations  on  filter  designs,  is  that  they  all 
refer  to  ideal  designs.  As  a  practical  application,  the  designs  are  seldom  used 
under  the  premises  of  the  design,  particularly  with  respect  to  impedance  matching. 

MR.  DEPUE:     How  was  the  laboratory  work  checked  up  on  these  tests? 

MR.  DURST:  Regular  commercial  prints  were  used  for  listening  tests.  Fre- 
quency measurements  were  taken  with  calibrated  constant-frequency  films 
and  a  warble  film;  also,  oscillators  were  coupled  to  the  input  of  the  amplifier 
systems. 

In  the  last  illustration  you  will  notice  the  absence  of  this  dip  of  250  cycles, 
that  is,  to  the  same  extent  as  it  appears  in  the  other  illustrations.  This  happens 
to  be  a  studio  listening  room.  The  point  I  wish  to  bring  out  in  this  connection 
is  that  I  believe  it  is  important  to  the  best  interests  of  all  concerned  that  the  manu- 
facturers of  theater  equipment  know  what  type  of  characteristic  the  producers 
are  using  in  listening  to  their  products.  If  we  ever  hope  to  make  it  possible  for 
theater  customers  to  hear  the  same  things  that  the  producers  hear  in  their  review 
rooms,  we  must  have  the  same  type  of  characteristic.  The  example  given  shows  a 
decided  difference  from  measured  theater  characteristics.  If  that  is  a  desired 
characteristic  or  a  necessary  one  for  recording  purposes,  then  all  theaters  should 
have  the  same.  However,  we  find  it  quite  the  contrary  in  reproducing  a  wide 
number  of  film  products.  Filling  in  this  dip  does  not  give  a  most  pleasing  result. 

MR.  HOVEY:  Is  that  caused  by  acoustic  conditions  in  the  theater?  I  should 
think  the  first  step  would  be  to  correct  that. 

MR.  DURST:  The  necessity  of  a  dip  in  the  region  of  250  cycles  is  general, 
because  of  variations  in  recording  practice  with  respect  to  dialog  equalization. 
It  is  usually  created  in  the  reproducing  characteristic  by  impedance  mismatching 
or  in  the  design  of  the  low-frequency  horns.  It  may  be  corrected  by  acoustical 
networks  within  certain  limits. 

*  With  respect  to  Figs.  14,  15,  and  16,  appreciation  for  his  cooperation  is  ex- 
pressed to  Mr.  L.  Ryder,  Director  of  Recording,  Paramount  Pictures,  Inc.,  Holly- 
wood, Calif. 


Feb.,  1939] 


FILM  REPRODUCER  SYSTEMS 


185 


MR.  FRIEDL:  It  can  also  be  done  electrically  with  more  flexibility,  pro- 
vided we  agree  on  what  we  are  trying  to  achieve.  There  are  several  ways  of 
doing  it.  We  are  all  striving  for  uniformity  in  the  presentation  of  our  product. 


teumve  SOUND  wrfNsirr  /N  OB 
•  i  t  i  i  + 

01  8  5;  S  o,  o  o> 

/ 

^ 

- 

N 

\+ 

J 

f 

^ 

^ 

^ 

\  , 

/ 

^ 

/ 

\ 

s 

^-^ 

N 

N, 

V 

\ 

RCA  #MBL£  fffeQ.  FILM  #^)e-l 
GfN.  RADIO  SOUND  L£V£L  M£7£R  #7S*A 

\ 

^ 

3O 


/OO 


500  fffC 

F#eOV£NCY /N  CYCLfS  P£ff  SECOND 


(OKC 


FIG.  15.     Overall  system  acoustic  characteristic  (Paramount  Studio  Sound 
Theater) .     Tests  made  by  the  Sound  Department  of  Paramount  Studio. 

If  we  agree  on  what  we  want  in  the  end,  the  studios  will  work  cooperatively,  and 
we  should  achieve  the  uniformity  we  desire. 

MR.  HOVEY  :     The  statement  was  made  several  times  that  this  or  that  sound 
was  satisfactory  or  good.     I  can  not  help  wondering  to  what  extent  a  state  of 


-. 


S~\ 

^ 

V 

/ 

** 

N 

^ 

•NS 

s 

. 

/ 

XX 

\ 

^ 

K 

s 

\ 

1 

RCA  WAR5L£  FK£O.  FILM  *238-f 
C/M  RAD/O  SOUND  LfV£L  M£Te##759-A 

V 

*—•* 

N 

^ 

30 


fOO  3CO  fKC 

FK£QU£MCY  IN  CYCLES  Pffi  SECOND 


fOffC 


FIG.  16.     Overall  system  acoustic  characteristic  (Paramount  Studio  Projection 
Room) .     Tests  made  by  the  Sound  Department  of  Paramount  Studios. 


tone  paralysis  may  enter  into  that.  It  has  been  my  experience  that  when  an  in- 
stallation is  made,  those  who  are  hi  contact  with  it  daily  become,  after  a  month  or 
so,  so  sound  paralyzed  that  even  the  most  terrible  sound  seems  all  right  to  them. 
I  wonder  whether  that  is  the  experience  of  other  engineers. 


186  F.  DURST  AND  E.  J.  SHORTT  [j.  s.  M.  p.  E. 

MR.  FRANK:  I  might  explain  that  by  a  story  I  have  heard  often  about  the 
Board  of  Directors  of  the  old  Victor  Talking  Machine  Company.  When  the  or- 
thophonic  phonograph  was  demonstrated  to  them,  they  said,  "That  is  wonderful, 
but  it  does  not  sound  like  a  phonograph."  It  certainly  has  been  our  experience 
in  theater  work  that  the  longer  we  listen  to  sound  quality,  irrespective  of  what 
the  quality  is,  the  better  we  think  it  is,  and  it  is  only  by  direct  comparison  and 
technical  analysis  that  we  begin  to  tell  whether  we  are  hearing  what  is  supposed  to 
be  "good"  sound  or  not. 

MR.  HOVEY:  Recently  I  witnessed  an  installation  of  what  I  considered  un- 
usually good  sound,  and  the  theater  manager  ordered  it  out  and  replaced  it  with 
an  outfit  about  five  years  old  because,  he  said,  it  sounded  better.  It  would  be 
interesting  to  know  whether  that  is  an  unusual  case  or  whether  other  engineers  run 
into  the  same  thing. 

MR.  GOLDSMITH:  That  is  an  unfortunately  frequent  case.  In  the  early  days 
of  radio,  there  was  once  a  controversy  between  the  advocates  of  long  horns  and 
short  horns,  respectively,  for  loud  speakers,  the  "long  horns"  stating  that  the 
loud  booming  response  with  practically  no  high  frequency  was  admirable  and  that 
it  was  what  they  termed  mellow  and  soothing;  whereas  the  other  group,  the 
"short  horns,"  insisted  on  suppression  of  low  frequencies  and  emphasis  on  high 
frequencies  until  the  result  was  a  thin  squeak.  No  agreement  was  ever  reached 
between  those  two  groups  because  each  became  confirmed  in  its  conviction  that 
its  customary  preference  constituted  ideal  reproduction.  There  was  no  major 
progress  until  response  curves  were  used  as  a  guide  to  design. 

MR.  FRIEDL:  Another  analogy  might  be  that  although  there  are  some  really 
good  radios  on  the  market  today  and  high-quality  radio  programs  are  broadcast, 
the  reproduction  in  the  average  home  is  poor  because  the  sound  is  unbalanced  by 
inaccurate  adjustment  of  the  tone  control. 

We  all  desire  to  standardize  theater  equipment  for  high-quality  reproduction. 
The  companies  interested  in  the  production  of  films  spend  a  lot  of  money  to  this 
end.  It  is  their  desire  to  control  the  quality  without  recommending  the  use  of 
variable  adjustments  that  would  permit  the  projectionist  to  distort  the  balance 
of  the  frequency  range  with  an  adjustable  tone  control.  Unless  we  can  agree 
on  a  certain  quality  of  performance  and  consistently  control  the  variables  that 
would  disturb  that  quality,  we  are  forced  to  admit  that  "good  quality"  is  a  matter 
of  personal  judgment.  Thus:  If  a  man  who  is  running  that  theater  wants  to 
have  it  sound  a  certain  way,  a  way  that  he  feels  satisfies  his  patrons,  he  will  de- 
mand the  same  facilities  as  he  uses  in  his  home  with  his  radio.  After  all,  in  the 
dubbing  of  a  film  there  is  a  man  who  sits  in  a  room  where  he  listens  and  judges 
what  he  thinks  is  good  reproduction.  He  might  satisfy  the  directors;  he  might 
satisfy  the  producer  and  the  studio  personnel  that  he  has  done  a  good  job.  At 
the  same  time  the  deadline  of  the  picture  might  be  approaching  or  the  budget 
running  out,  and  that  picture  is  going  out  to  make  money.  The  people  in  the 
theater  end  of  the  chain  are  supposed  to  correct  all  those  ills.  We  hope  the  studio 
recording  can  be  perfected  to  the  point  where  we  can  eliminate  that  variable. 

MR.  WOLF:  The  only  criticism  I  would  have  to  the  overall  response  curve  is 
that  it  shows  too  many  factors,  represents  so  many  things  that  you  do  not  know 
what  is  at  fault. 


Feb.,  1939]  FlLM  REPRODUCER  SYSTEMS  187 

Where  do  you  think  the  weakest  link  now  exists?  Is  it  in  the  theater  acoustics 
or  in  the  electrical  system?  Or  is  it  still  in  the  loud  speaker  system? 

MR.  DURST:  I  believe  that  by  careful  manipulation  of  the  characteristics,  a 
given  theater  can  be  made  to  sound  fairly  satisfactory.  Granted  there  are  acousti- 
cal conditions  that  can  not  be  compensated  for  electrically,  as  a  general  average  one 
can  make  equipment  sound  right.  But,  if  what  is  put  on  the  film  is  not  designed 
to  be  reproduced  over  that  characteristic,  it  never  will  sound  right.  If  a  given 
theater  is  adjusted  so  that  it  sounds  right  for  one  film,  the  next  one  that  comes 
into  that  house  may  be  altogether  different,  and  the  service  engineer  has  to  go 
back  and  readjust  or  re  tune  the  entire  system  to  satisfy  his  customers. 

MR.  WOLF:  I  still  believe  in  the  theory  of  the  uniform  characteristic  in  every 
element  of  the  circuit  from  the  microphone  all  the  way  through.  A  great  many 
are  getting  away  from  that,  thinking  perhaps  they  can  have  compensating  in- 
fluences in  each  element  of  the  system,  but  I  still  think  we  will  not  get  a  uniform 
response  until  in  the  theater,  we  have  uniformity  in  every  piece  of  equipment. 

MR.  DURST:  I  do  not  believe  that  is  altogether  desirable  from  the  recording 
standpoint.  Producers  have  such  diversity  of  story  material,  and  in  their  effort 
to  reproduce  the  true  dramatic  effect,  it  would  be  rather  difficult  for  them  to  have 
any  fixed  characterics,  but  I  do  believe  that  they  must  all  work  to  an  established 
end  result.  In  like  manner,  I  think  the  manufacturer  of  reproducing  equipment 
should  work  to  that  same  end  result.  It  should  be  up  to  him  to  decide  whether 
he  wants  to  taper  off  his  horn  or  whether  he  wants  to  change  his  slit  size.  If  he 
can  obtain  a  better  signal-to-noise  ratio  and  better  overall  compensation  by  one 
means  or  the  other,  I  think  it  should  be  left  to  his  discretion. 

MR.  STROCK:  I  should  like  to  say  one  word  about  all  these  variables  that  enter 
into  the  problem.  In  our  own  particular  case,  and  I  know  it  to  be  true  in  several 
of  the  other  studios,  what  we  do  shows  up  only  in  what  comes  out  of  the  horn 
in  the  theater,  and  we  are  continually  checking  the  product  that  we  listen  to  in  our 
own  theater  from  day  to  day  against  what  it  sounds  like  in  the  field.  Of  course, 
you  have  to  define  a  "representative  theater,"  but  nevertheless  it  is  a  good  theater 
that  is  generally  accepted  as  giving  good  sound.  After  all,  when  you  come  down 
to  saying  whether  sound  is  good  or  whether  it  is  bad,  our  own  definition  of  good 
sound  is  sound  that  comes  out  of  a  loud  speaker  that  is  natural  and  tempered  by 
the  perspective  of  what  is  going  on  in  the  picture;  meaning  that  if  you  have  a  full 
head  close-up  of  somebody  saying  some  very  touching  and  endearing  words,  you 
have  to  temper  that  by  the  size  of  the  picture  and  the  dramatics  of  the  story. 
Nevertheless,  good  sound  is  natural  sound. 

MR.  WOLF  :     What  you  are  after  is  a  facsimile  of  the  original  in  most  cases. 

MR.  STROCK:     Not  forgetting  the  picture. 


SOME  PRACTICAL  ACCESSORIES  FOR  MOTION 
PICTURE  RECORDING* 


R.  O.  STROCK** 


Summary. — The  addition  of  practical  operational  accessories  to  standard  recording 
channels  as  purchased  expedites  operation  and  saves  time.  At  the  Eastern  Service 
Studios  a  number  of  such  accessories  have  been  designed  and  are  described  briefly. 
It  is  the  purpose  of  this  paper  to  show  what  has  been  done  at  one  studio  in  the  hope  that 
it  may  be  of  some  interest  and  help  to  others  who  are  engaged  in  recording  work. 

Included  in  the  equipment  are  the  following  items:  A  small  collapsible,  portable 
microphone  boom  for  location  work;  a  special  microphone  suspension  to  prevent  me- 
chanical noises  from  getting  into  the  recording  system;  a  small  mixer  console  for  stage 
work,  to  permit  the  mixer  man  to  operate  close  to  the  scene  of  action;  an  accurate 
illumination  meter,  using  a  microammeter,  for  setting  and  checking  the  recording 
machine  exposure;  a  compact  re-recording  mixer  console  equipped  with  equalizers, 
effect  filters,  amplifiers,  and  attenuators;  a  projected  volume  indicator  and  footage 
counter  for  use  in  re-recording  rooms;  a  film  playback  adapter  for  use  on  a  Western 
Electric  film  machine  for  location  use;  playback  horns  for  stage  and  location  use; 
and  an  air-brush  adaptation  for  blooping  re-recording  tracks. 

When  recording  channels  are  purchased  they  usually  consist  of 
several  separate  units  following  the  general  order  and  layout  of  the 
electrical  schematic.  The  addition  of  certain  practical  operational 
accessories  to  these  standard  recording  channels  expedites  operation 
and  saves  time.  At  the  Eastern  Service  Studios  a  number  of  such 
accessories  have  been  designed  which  are  briefly  described  and  illus- 
trated herein.  It  is  the  purpose  of  this  paper  to  show  what  has  been 
done  in  this  studio  in  the  hope  that  it  may  be  of  some  interest  and 
help  to  others  engaged  in  recording. 

Portable  Microphone  Booms. — The  first  unit  to  be  described  is  a 
small  portable,  collapsible  microphone  boom.  For  regular  studio  use 
we  have  the  standard,  medium,  and  large  types  of  Mole-Richardson 
booms.  There  are  many  instances  where  a  small  microphone  suspen- 
sion is  needed  such  as  in  pick-ups  from  a  theater  stage,  small  attic 
rooms,  narrow  hallways,  etc.  We  have  two  types,  one  a  very  simple 

*  Presented  at  the  1938  Fall  Meeting  at  Detroit,  Mich. ;  received  October  17, 
1938. 

**  Eastern  Service  Studios,  Long  Island  City,  N.  Y. 

188 


SOME  ACCESSORIES  FOR  RECORDING 


189 


190 


R.  O.  STROCK 


[J.  S.  M.  P.  E. 


one  and  the  other  a  bit  more  versatile  but  not  complicated.  The 
smaller  one  (Fig.  1)  consists  merely  of  two  telescoping  duralumin  rods 
mounted  on  a  yoke  which  is  equipped  to  fit  in  a  standard  light  stand. 

This  is,  of  course,  useful  only  for 
stationary  shots.  The  general 
purpose  small  boom  is  shown  in 
Fig.  2.  It  may  be  extended  or 
retracted,  lowered  or  raised  noise- 
lessly at  will,  and  it  is  easily 
moved  for  trucking  shots.  It 
may  be  locked  in  any  position  by 
friction  by  the  handle.  It  is 
easily  dismantled  for  packing  in 
the  location  trucks.  Its  total 
weight  is  about  fifty  pounds. 

Microphone  Suspensions. — 
Most  microphones  are  subject 
to  the  disadvantage  of  convert- 
ing mechanical  shocks  into  elec- 
trical noises  in  the  recording  sys- 
tems. The  standard  mountings 
(Fig.  3)  for  either  the  Western 
Electric  630  or  618  microphones 
do  not  provide  for  any  shock 
absorbing  medium  between  the 
microphone  and  the  boom  or 
support.  Such  an  absorbing 
medium  is  necessary  when  mov- 
ing the  microphone  rapidly  dur- 
ing the  recording.  Microphone 
noise  is  largely  due  to  mechani- 
cal noises  within  the  boom  being 
transmitted  directly  into  the 
microphone.  To  help  eliminate 
this  direct  transmission,  a  short 
flexible  lead  (Fig.  4)  is  provided 
between  the  boom  cable  and  the  microphone.  This  connector  is 
usually  made  from  telephone  tinsel  or  other  very  flexible  stranded 
wire  and  has  proved  very  satisfactory.  A  retaining  cord  must  be 
provided  to  avoid  excess  strain  on  the  fragile  tinsel  leads. 


FIG.  3.      ( Upper]  Standard  microphone 
mounting. 

FIG.  4.  (Lou'er)  Showing  flexible 
lead  between  microphone  boom  cable 
and  the  microphone. 


Feb.,  1939] 


SOME  ACCESSORIES  FOR  RECORDING 


191 


The  microphones  are  supported  in  several  different  types  of  mount- 
ings all  of  which  are  satisfactory  from  a  noise  standpoint.  Shown  in 
Fig.  4  is  the  mounting  for  the  630  microphone.  Note  that  the  mi- 
crophone is  held  in  a  yoke  which  is  supported  by  standard  Lord  rub- 
ber mountings  for  reducing  mechanical  shock  to  the  microphone 


FIG.  5.     (Upper)  Mounting  for  618  microphone. 

FIG.    6.     (Lower)    Mounting    for    618    or    630 

microphone. 


proper.  In  Fig.  5  is  the  mounting  for  the  618  microphone.  In  this 
mounting  the  shock  is  absorbed  by  supporting  the  microphone  in  a 
ring  held  by  elastic.  In  Fig.  6  is  shown  a  mounting  which  may  be 
adapted  to  either  the  618  or  the  630  microphone.  All  these  mount- 
ings, when  used  in  conjunction  with  the  flexible  connector,  prevent 
mechanical  noise  from  being  transferred  to  the  electrical  recording 
system. 


192  R.  O.  STROCK  [j.  s.  M.  P.  E. 

Mixer  Consoles. — Most  present-day  motion  picture  voice  recording 
is  monitored  and  mixed  from  a  small  mixing  console  which  can  be 
placed  close  to  the  set  or  scene  of  action,  as  shown  in  Fig.  7.  The 
advantages  of  the  mixer  man's  being  in  close  contact  with  the  director 
and  the  cameraman  and  in  such  a  position  that  he  can  at  all  times 
watch  the  action  far  outweighs  the  disadvantages,  if  any,  of  head- 
phone monitoring.  No  loud  speaker  in  a  small  monitor  booth  can 
equal  theater  characteristics.  The  mixer  man  must  make  a  personal 
judgment  between  what  he  is  hearing  when  recording  and  what  he 


FIG.  7.     Small  mixing  console  on  set. 

will  hear  in  the  finished  sound  in  a  theater.  Therefore,  he  might  just 
as  well  make  his  judgment  between  what  he  hears  in  the  loud  speaker 
compared  to  the  theater  horns  as  between  a  booth  loud  speaker  and 
the  theater  horns. 

Mixer  consoles,  as  used  at  Eastern  Service  Studios,  are  as  shown  in 
Fig.  7.  They  are  small  and  can  be  moved  easily  when  the  company 
moves  to  another  set.  A  three-position  mixer  is  used  and  has  been 
found  adequate  for  most  ordinary  picture  work.  The  mixer  is  con- 
nected to  an  amplifier,  and  the  combination  is  known  as  an  ERPI  RA- 
150,  which  is  battery  operated  and  uses  electronic  mixing. 


Feb.,  1939] 


SOME  ACCESSORIES  FOR  RECORDING 


193 


The  mixer  is  so  connected  that  if  it  is  necessary  to  move  quickly  to 
a  nearby  set  for  a  pick-up  shot,  the  mixer  panel  can  be  easily  removed 
and  carried  to  the  location  without  moving  the  console.  It  is  pos- 
sible to  work  the  mixer  300  feet  away  from  its  amplifier. 

On  the  panel  board  (Fig.  7)  are  shown  the  mixer,  the  telephone  for 
interstage  communication,  and  the  signal  lights  for  the  recording 
system.  The  amplifier  is  housed  in  one  end  of  the  console  and  the 
battery  equipment  in  the  other.  In  the  rear  is  space  for  storing  the 
microphone  cable. 


FIG.  8.     General  layout  of  re-recording  mixer 
console. 


Projected  Volume  Indicator  and  Footage  Counter  for  Use  in  Re-Re- 
cording.—The  next  unit  to  be  described  is  the  re-recording  mixer  con- 
sole. This  is  very  similar  to  the  stage  pick-up  unit  but  is  used  for  the 
re-recording  process  and  is  a  bit  more  complicated.  Fig.  8  shows  its 
general  layout.  The  circuits  are  brought  from  the  re-recording 
machines  through  the  jack  field  on  the  end  of  the  table  and  then 
through  the  mixer  and  into  a  main  amplifier.  The  amplifier  is  a-c. 
operated  and  is  mounted  on  a  swinging  hinge  so  it  may  be  swung  out- 
ward for  changing  tubes  or  for  servicing.  Equalizers,  attenuators, 
telephone  effect  filters,  high-  and  low-pass  cut-off  filters,  and  a  uni- 
versal high-  and  low-frequency  equalizer  are  provided  so  they  may 
be  inserted  in  any  desired  mixer  position  for  changing  the  circuit  fre- 


194  R.  O.  STROCK  [j.  s.  M.  p.  E. 

quency  characteristics.     A  level  control  is  provided  on  the  output  of 
the  amplifier  for  controlling  general  level  into  the  recording  rooms. 
Re-recording  mixer  men  must  first  be  artists,  and  second  engineers, 


FIG.  9.     Projected  volume  indicator  and  footage  counter. 


FIG.  10.     Showing  arrangement  of  volume  indicator  and 
footage  counter. 

for  in  the  mixing  of  many  sound-tracks  into  a  composite  effect,  recog- 
nition must  be  made  of  cueing,  levels,  artistic  effect  required  by  the 
director,  perspective,  and  geography  of  the  scene  at  hand.  In  order 


Feb.,  1939]  SOME  ACCESSORIES  FOR  RECORDING  195 

to  aid  the  mixer  man  in  doing  so  many  things  at  one  time,  and  in  addi- 
tion, not  divert  his  attention  from  the  picture,  a  projected  volume  in- 
dicator and  footage  counter  has  been  built.  As  can  be  seen  in  Fig.  9, 


FIG.  11.    Footage  counter. 

the  volume  indicator  and  footage  counter  images  are  projected  to  a 

considerable  size  directly  below  the  picture,  so  it  is  possible  for  the 

mixer  to  see  with  ease  the  footage  for  spotting  cues,  the  sound  volume, 

and  the  picture,  all  at  the  same  time. 

If  the  volume  indicator  is  used  on  the 

re-recording  table  it  is  very  difficult  and 

tiring  to  try  to  change  the  line  of  sight 

between   the  picture   and    the   volume 

indicator  rapidly.     This  unit  has  been  a 

great  help  in  re-recording. 

Fig.  10  shows  how  simply  this  has 
been  accomplished.  The  volume  indica- 
tor is  placed  on  the  rear  of  a  standard 
Keystone  postcard  projector  and  its 
image  projected  upon  the  screen.  Two 
100- watt  lamps  are  used  to  illuminate 
the  meter  and  no  difficulty  from  heating 
is  experienced,  the  lamps  remaining  on 

for  several  days  at  a  time  during  long 

J  6        FIG  12.     Illumination  meter, 

re-recording  sessions. 

The  footage  counter  also  is  shown  in  Fig.  10.     A  special  Veeder 
counter,  with  the  numbers  upside  down,  was  mounted  on  an  inter- 
locked motor  which  is  electrically  connected  to  the  recording  dis- 


196 


R.  O.  STROCK 


[J.  S.  M.  P.  E. 


FIG.  13.     Recording  machine  with  illumination  meter  attached. 


FIG.  14.     Recording  machine  before  being  adapted  for 
playback. 


Feb.,  1939] 


SOME  ACCESSORIES  FOR  RECORDING 


197 


tributor  system  the  same  as  the  recording  and  projection  motors. 
The  counter  (Fig.  11)  is  illuminated  by  a  100- watt  screw-base  lamp 
with  a  condenser  lens,  and  its  image  is  projected  upon  a  screen  directly 


FIG.  15.     Playback  adapter. 


FIG.  16.     Playback  adapter  unit  mounted  on  the  machine. 

below  the  volume  indicator  image  (Fig.  10).  The  counter  is  easily 
reset  and  the  volume  indicator  meter  can  be  removed  easily  for  re- 
placement if  necessary. 

Illumination  Meter  for  Checking  Recording  Machine  Exposure. — In 
order  to  control  the  quality  of  sound  recording  it  is  necessary  to  main- 


198 


R.  O.  STROCK 


(J.  vS.  M.  P.  E. 


tain  the  recording  machine  exposures  very  accurately.  Ordinary 
ammeters  in  the  lamp  circuit  can  not  be  read  accurately  enough  to 
control  the  illumination  within  the  necessary  limits.  Our  standards 

require  that  the  negative  expo- 
sure be  held  within  0.05  in 
visual  diffuse  density.  In  order 
to  hold  the  exposure  within 
this  rather  narrow  limit  an  ac- 
curate illumination  meter  was 
designed,  as  shown  in  Fig.  12. 
It  consists  of  a  plate  which  can 
be  clamped  quickly  on  the  re- 
cording machine  (Fig.  13).  On 
the  plate  is  mounted  a  metal- 
lic mirror  which  intercepts  the 
light-beam  just  in  front  of  the  film  and  reflects  the  light  into  a 
photoelectric  cell  mounted  in  the  round  container  on  the  front 
of  the  plate.  The  photoelectric  cell  has  low  sensitivity  and 


FIG.  17.     Blooper. 


FIG.  18.     Another  design  of  blooper. 


is  operated  in  the  very  stable  portion  of  its  sensitivity  curve.  It  is 
connected  to  a  0-2jia  Rawson  microammeter  and  a  normal  current  of 
approximately  0.75  microampere  flows  through  the  photoelectric  cell. 
An  exposure  test  is  made  for  the  particular  emulsion  in  use  and  the 


Feb.,  1939]  SOME  ACCESSORIES  FOR  RECORDING  199 

correct  density  is  determined  in  the  usual  manner  from  the  negative 
H&D  curve,  after  which  a  curve  is  made  plotting  density  against 
microammeter  readings,  and  the  correct  reading  obtained.  The  il- 
lumination can  be  checked  very  rapidly  by  the  use  of  this  instrument 
and  it  is  the  usual  procedure  to  check  the  exposure  after  every  five 
takes.  The  instrument  has  been  in  use  for  several  years  and  has 
proved  very  satisfactory,  showing  that  the  exposure  can  be  main- 
tained within  the  0.05  density  limits  for  months  at  a  time. 

Film  Playbacks  for  Location  Trucks. — It  is  often  necessary  to  use 
film  playbacks  on  location.  The  standard  D 867 15  Western  Electric 
recording  machine  for  location  trucks  is  not  equipped  for  playbacks. 
One  of  our  machines  in  a  location  truck  has  been  modified  so  that  it 
can  be  used  for  either  recording  or  playback  purposes.  In  Fig.  14 
is  shown  the  recording  machine  before  adapting  it  to  playback.  In 
Fig.  15  is  shown  the  adapter.  It  consists  of  two  arms  for  holding  the 
supply  and  take-up  reels  and  a  means  of  driving  it.  In  Fig.  16  the 
unit  is  shown  mounted  on  the  machine.  A  photoelectric  cell  is  placed 
inside  the  recording  sprocket  and  its  __„____— _~ 

output  fed  through  a  low-capacity 
cable  to  the  regular  split-beam 
monitoring  unit,  whose  output  is 
fed  through  the  recording  amplifiers 
and  then  into  a  playback  horn.  The 
unit  can  be  removed  and  the  ma- 
chine returned  to  the  recording  FlG- 19-  Blooper  of  Fig  18  with 

plate  in  place  over  film, 
condition  in  only  a  few  minutes  by 

removing  two  thumb-screws  and  removing  the  drive  belt.  The  film 
can  be  rewound  without  removing  the  reels  from  the  machine. 

Air-Brush  for  Blooping. — In  re-recording  it  is  very  necessary  that 
bloops  caused  by  splices  be  entirely  eliminated.  We  use  two  types  of 
"bloopers."  One  is  as  shown  in  Fig.  17,  described  by  E.  I.  Sponable 
some  time  ago.  This  unit  does  the  job  very  well.  We  have  designed 
another  unit  that  does  the  job  equally  well  and  is  shown  in  Fig.  18. 
It  consists  of  a  block  and  template  with  centering  pins  and  an  air- 
brush for  spraying  the  blooping  ink  on  the  template  and  the  splice. 
Rapid  drying  ink  is  used.  In  Fig.  19  is  shown  the  plate  in  place  over 
the  film.  The  template  provides  assurance  that  the  edges  of  the 
bloop  are  sharp  and,  because  the  back  of  the  film  is  pressed  against 
rubber,  good  contact  can  be  obtained  between  the  template  and  the 
film.  The  design  of  the  bloop  patch  is  the  same  as  the  one  described 


200  R.  O.  STROCK 

by  Sponable.  This  unit  can  be  constructed  very  easily,  and  a  stand- 
ard air-brush  and  tank  are  used. 

The  accessories  described  above  have  been  a  great  help  to  this 
studio  in  its  operation.  It  is  hoped  that  others  engaged  in  recording 
work  will  describe  their  many  operational  accessories  from  time  to 
time  in  the  JOURNAL. 

Credit  for  the  design  and  suggestions  on  the  units  just  described  is 
given  to  our  recording  staff  in  general  and  in  particular  to  Dan  Don- 
caster,  our  mechanic,  who  built  and  suggested  many  of  the  items. 


THE  LIGHTING  OF  MOTION  PICTURE  THEATER 
AUDITORIUMS* 


F.  M.  FALGE  AND  W.  D.  RIDDLE** 


Summary. — Here  and  there  a  theater  is  planned  with  lighting  features  utilizing  the 
fundamental  principles  that  have  been  expounded  on  many  occasions.  In  too  many 
cases,  however,  interior  lighting  has  lagged  far  behind  exterior  lighting  for  advertising, 
and  owner  and  public  alike  have  suffered.  In  too  many  cases,  also,  the  theater  falls 
far  short  of  complementing  the  attractive  scenes  so  well  projected  upon  the  screen. 

This  paper  reiterates  the  aims  and  advantages  of  proper  lighting,  and  outlines  the 
problem  of  locating,  and  controlling  the  lighting  properly  so  that  it  will  be  comfort- 
able and  pleasing  and  an  aid,  psychologically. 

Because  of  the  almost  infinite  variation  in  design  for  theater  audi- 
toriums, with  influences  all  the  way  from  cave-dwellers  to  the  ultra- 
modern, and  from  the  bottom  of  the  sea  to  the  sky  above,  practically 
every  conceivable  lighting  method  or  idea  has  been  called  into  play. 
Unquestionably  architectural  influence  and  decorative  character  have 
played  a  far  more  important  part  than  the  provision  of  light  for  com- 
fortable and  safe  seeing. 

The  purposes  of  auditorium  lighting  are  several  and  varied.  In 
this  investigation  we  are  chiefly  concerned  from  the  viewpoint  of  com- 
fortable vision  and  how  to  provide  for  it  while  serving  these  other  re- 
quirements as  well.  This  paper  presents  experimental  data  applying 
to  a  limited  range  of  auditorium  conditions.  Although  by  no  means  a 
comprehensive  treatment,  it  does  offer  a  little  additional  information 
in  a  field  where  quantitative  studies  have  been  badly  needed. 

In  order  to  visualize  the  complete  picture  of  the  objectives  of  audi- 
torium lighting  as  set  forth  on  previous  occasions li2'3-4  they  are  re- 
peated here : 

(1)  Comfortable  Vision. — Eyes  should  be  aided  in  their  adjustment  to  darkened 
interiors  and  made  comfortable  by  adhering  to  brightness  standards. 

(2)  Convenience. — People  must  see  quickly  and  easily  to  locate  seats,  without 
annoyance  to  others  and  without  individual  usher  service  which  is  expensive. 

*Presented  at  the  1938  Fall  Meeting  at  Detroit,  Mich.;   received  October 
19,  1938. 
**General  Electric  Company,  Cleveland,  Ohio. 

201 


202 


F.  M.  FALGE  AND  W.  D.  RIDDLE 


[J.  S.  M.  P.  E. 


(3)  Safety. — Light  dispels  the  fears  that  patrons  may  feel  in  darkened  theaters. 
Accidents  and  accident  costs  are  also  reduced. 

(4)  Program  Appreciation. — Decorative  lighting  provides  a  "plus"  value  that 
secures  a  favorable  psychological  reaction,  and  is  of  aid  in  counteracting  seasonal 
temperature  complexes.     It  aids  special  programs  or  holiday  celebrations  and 
creates  a  mood  that  aids  program  appreciation.    It  makes  people  look  better  and 
feel  better. 

(5)  Cleanliness. — Light  reveals  a  clean  theater. 

As  previously  stated,  it  is  desirable  to  give  consideration  to  all 
these  factors  in  lighting  an  auditorium,  but  this  paper  will  deal  pri- 
marily with  the  one  factor  of  ocular  comfort  as  affected  by  the  quan- 
tity and  direction  of  light  and  the  location  of  sources,  their  brightness 
and  that  of  the  several  parts  of  the  visual  field. 


c            o 

i  —  ii  —  ii  —  ii  —  ii  —  inn 

"f 

LUMINOUS  W.NOOWS                        > 

\\ 

8 

B 

L______ 

SCREEN 

PROSCENIUM  PLAN 

SIOEWALL-PLAN 

1- 

46-0 

—  H 

SEC 

°%ffi£ 

BSE 
I 

»vt 

n 

f 

DOWNLIOHTS                    M 

oo        «r      o       >>       o 

' 

J 

-/ 

; 

0 

BWCHTNCSS  MEASUREMENT! 
POSITIONS  AT  SEATS 

/         ^ 

SCREEN 

^^,INOIRECT  LUMINAIRES. 

O     Q^       o          13           0*0 

SCREEN 

,/ 

n 

N  -BOTTOM 

-ICURTAW' 

*P»INCI«_E   OBSERVER  POSITION                         |  I 

| 

FLOOR-PLAN  CEILING-PLAN 

FIG.  1.     Auditorium  layout  with  illumination  systems  used  in  tests.     Letters 
refer  to  locations  measured  for  brightness. 

In  the  past  we  were  handicapped  by  the  lack  of  a  convenient  and  in- 
expensive brightness  meter  that  would  permit  ready  recording  and 
analysis  of  brightness  conditions  throughout  an  auditorium  as  viewed 
from  any  given  seat.  Some  years  ago  the  SMPE  Theater  Lighting 
Committee10  devoted  much  study  to  the  characteristics  of  such  an 
instrument.  This  need  has  been  expressed  in  many  Committee  re- 
ports from  time  to  time.  Today,  this  need  is  met  by  the  Luckiesh- 
Taylor  brightness  meter.5  It  is  possible  to  focus  the  instrument  on  a 
very  small  spot,  and  accurately  determine  the  brightness  of  that  spot 
over  a  range  exceeding  that  found  in  theaters.  The  brightness  is 
read  directly  in  candle-power  per  square-inch,  or  in  the  more  readily 
comprehended  foot-lambert. 


Feb.,  1939]  LIGHTING  THEATER  AUDITORIUMS  203 

The  foot-lambert  may  be  defined  simply  as  the  brightness  of  a 
perfect  diffuser  emitting  one  lumen  per  square-foot.  Or,  assuming 
perfect  diffusion,  the  foot-lambert  is  the  brightness  resulting  from  the 
product  of  foot-candles  illuminating  a  surface  multiplied  by  the  re- 
flection or  transmission  factor  of  the  surface.  Hence  it  is  sometimes 
referred  to  as  foot-candle  on  white. 

The  tests  conducted  by  the  authors  were  made  with  six  experienced 
observers  in  the  auditorium  of  the  Lighting  Institute  of  Nela  Park. 
The  dimensions  of  this  auditorium  are  46  feet  by  34  feet  and  the  seat- 
ing capacity  is  250.  Fig.  1  illustrates  the  auditorium  layout  and  loca- 
tion of  test  positions.  The  small  size  of  the  theater  will,  of  necessity, 
serve  as  a  limit  in  the  application  of  results  to  large  theaters. 

The  lighting  systems  used  in  the  tests  and  located  in  this  audi- 
torium have  adequate  thyratron  and  resistance  control.  They  in- 
clude : 

(1)     Six  suspended  indirect  type  luminaires. 

(2}  Six  luminous  windows  at  each  upper  side  wall  with  lighting  directed  toward 
the  front  part  of  the  auditorium. 

(3}  Six  pin-hole  objective-lens  downlights  directed  at  an  angle  of  20  degrees 
toward  the  front  of  the  auditorium. 

(4)  Stage  borderlighting  to  light  screen  surroundings. 

(5)  Aisle  lights. 

To  fix  the  test  conditions  as  much  as  possible,  tests  were  first  made 
with  a  slide  picture  having  light  and  dark  regions  fairly  evenly  dis- 
tributed. The  picture  was  71  inches  wide  by  51  inches  high.  The 
brightness  of  screen  without  picture  averaged  14  foot-lamberts  and 
the  whites  in  the  picture  averaged  9  foot-lamberts.  The  principal 
test  position  was  a  side  seat  in  the  rear  row  which  offered  the  most 
severe  visual  condition  as  far  as  ceiling  and  side-wall  brightness  was 
concerned. 

It  is  readily  apparent  that  a  still  picture  would  become  monotonous 
and  distractions  would  be  more  in  evidence  than  with  a  moving  pic- 
ture. Some  of  the  tests  were  accordingly  re-run  with  a  motion  pic- 
ture, with  a  height  of  46  inches  and  a  width  of  65  inches.  At  the 
principal  test  positions,  43  feet  from  the  screen,  the  angle  subtended 
by  the  screen  was  7  degrees.  The  average  brightness  of  the  blank 
screen  with  projector  lighted  and  shutter  operating  was  4  foot-lam- 
berts. With  a  picture  the  brightest  whites  were  3  foot-lamberts. 
The  small  size  and  low  brightness  of  the  screen  are  two  of  the  limita- 
tions mentioned. 


204  F.  M.  FALGE  AND  W.  D.  RIDDLE          [j.  s.  M.  P.  E. 

Experienced  observers  were  selected  and  the  objectives  made  known 
to  them  because  this  test  was  purely  a  visual  one.  It  was  interesting 
to  note,  however,  how  positive  reactions  were  to  visual  discomfort, 
how  close  the  reactions  of  various  observers  checked,  and  how  close 
an  agreement  was  had  to  earlier  tests  by  others. 

Test  No.  1 — Lighting  from  Picture  Alone. — (Fig.  2)  Observers  con- 
cluded that  a  lighted  screen  alone  was  definitely  uncomfortable  due 
to  glare,  and  very  fatiguing,  a  conclusion  supported  by  general  ex- 
perience.1-4'6 Aisle  lights  did  not  aid  appreciably  in  relieving  this 


FIG.  2.     Test  No.  1 :    Lighting  from  picture  alone. 

condition.  It  was  not  possible  in  this  test  to  evaluate  the  discomfort 
factor  of  entering  such  a  darkened  auditorium  from  regions  of  higher 
brightness,  but  experience  in  theaters  lighted  in  this  manner  indi- 
cates that  it  is  definitely  uncomfortable  and  annoying.  This  is  sub- 
stantiated also  by  previous  investigations.  In  this  case  it  is  usually 
necessary  to  accompany  patrons  to  seats  with  a  flashlight,  a  procedure 
that  is  a  rather  poor  seeing  compromise  and  an  expensive  one.  Aisle 
lights  were  of  some  help,  due  primarily  to  identification  of  aisles. 

It  was  noted  that  variations  in  screen  brightness  due  to  unequal 
distribution  of  blacks  and  whites  caused  wide  fluctuation  of  illumina- 
tion on  the  ceiling  and  side  walls  that  was  distinctly  annoying.  This 


Feb.,  1939]  LIGHTING  THEATER  AUDITORIUMS  205 

condition  is  accentuated  when  patrons  wear  light-colored  summer 
clothing. 

Test  No.  2 — Illumination  of  Screen  Surroundings. — Previous  in- 
vestigations have  concluded  that  screen  surroundings  should  pre- 
sent some  brightness.  Jones4  arrived  at  the  conclusion  that  the 
contrast  with  the  immediate  surroundings  should  be  less  than  1  to 
500  as  compared  with  the  brightest  parts  of  the  picture.  O'Brien 
and  Tuttle7  concluded  that  the  highest  desirable  brightness  of  the 
immediate  surroundings  is  0.8  foot-lambert  and  that  the  preferred 
brightness  lay  between  0.05  to  0.20.  This  provides  a  contrast  range  of 
Vioo  to  J/26  of  the  bright  parts  of  the  picture.  Wolf8  concluded  that  a 
border  brightness  of  0.047,  providing  a  ratio  of  1  to  100  with  the 
brightest  parts  of  the  screen,  was  desirable.  Schlanger6  proposed  a 
method  of  automatically  varying  this  brightness  with  the  picture. 

In  their  tests  the  authors  wished  to  determine  the  desirable  bright- 
ness of  the  field  surrounding  the  screen  rather  than  just  the  frame  for 
the  picture.  Observers  agreed  that  as  surrounding  illumination  of 
fairly  uniform  brightness  increased  glare  was  relieved.  With  the  slide 
on  the  screen,  a  brightness  of  surroundings  was  soon  reached  in  which 
it  was  felt  that  the  background  began  to  compete  with  the  screen  it- 
self for  attention.  The  unlighted  border  around  the  screen  first  be- 
came objectionable  because  of  its  extreme  contrast  with  both  screen 
and  surroundings.  When  the  screen  was  arranged  so  as  to  be  im- 
mediately adjacent  to  the  illuminated  background  this  condition 
was  relieved  and  a  higher  brightness  of  background  was  satisfactory. 
A  brightness  of  background  of  l/Z6  to  Vw  of  that  of  the  screen  was 
found  satisfactory.  However,  although  relieving  the  screen  glare, 
little  illumination  was  added  to  the  auditorium  to  provide  the  com- 
plete comfort  condition  desired. 

Test  No.  3 — Upper  Value  of  Desirable  Illumination. — It  was  felt 
that  this  upper  limit  might  be  set  as  the  point  at  which  the  move- 
ments of  other  observers  became  disturbing  and  that,  before  deter- 
mining the  quality  or  quantity  of  light  desirable  for  comfort,  it  was 
necessary  to  determine  this  upper  limit.  This  value  was  determined 
with  a  fairly  comfortable  system  of  lighting  combining  background 
illumination  and  indirect  illumination  from  the  suspended  luminaires. 
With  the  slide,  this  value  lay  between  0.1  and  0.2  foot-candle.  How- 
ever, the  animation  of  the  motion  pictures  tended  to  render  less  con- 
spicuous the  movements  of  others,  and  values  of  0.2  to  0.5  foot- 
candle  were  found  to  be  satisfactory. 


206  F.  M.  FALGE  AND  W.  D.  RIDDLE          [j.  s.  M.  P.  E. 

This  condition  was  affected  to  a  great  extent  by  the  color  of  clothes 
worn.  At  0.5  foot-candle,  medium  or  dark  clothes  were  not  readily 
noticed  but  light  clothes  were.  The  brightness  of  white  clothes  in  this 
case  was  0.35  foot-lambert.  It  can  be  concluded,  therefore,  that  in 
the  south,  and  during  the  summer  when  light-colored  clothes  are 
worn,  levels  of  illumination  should  be  somewhat  lower. 

Test  No.  4 — Lower  Value  of  Desirable  Illumination. — Determination 
of  this  lower  value  was  more  difficult  and  the  factor  of  eye  adapta- 
tion caused  by  entering  the  auditorium  from  regions  of  higher  bright- 
ness had  to  be  disregarded.  The  indirect  lighting  system  was  selected 


FIG.  3.     Test  No.  7:    Downlighting. 

for  this  purpose.  Starting  with  only  the  motion  picture  on  the  screen 
and  gradually  raising  the  level  of  illumination,  it  was  found  that  re- 
lief from  the  glare  of  the  screen  came  when  the  auditorium  illumina- 
tion ranged  from  0.05  to  0.1  foot-candle.  Measurements  were  made 
in  foot-lamberts  with  the  brightness  meter  and  the  corresponding 
illumination  values  were  calculated. 

Test  No.  5 — Indirect  Lighting  (Suspended  Luminaires). — With  the 
screen  surroundings  having  a  brightness  of  1/w  of  the  screen  whites, 
several  levels  of  indirect  lighting  were  investigated. 


Feb.,  1939] 


LIGHTING  THEATER  AUDITORIUMS 


207 


With  0.5  foot-candle  in  the  auditorium,  the  upper  limit  reached  in 
test  No.  3,  the  lighting  was  fairly  comfortable  but  with  the  following 
shortcomings : 

(1)  The  light  falling  on  the  screen  measured  0.35  foot-lambert,  which  was 
enough  to  harm  the  picture  contrasts  seriously. 

(2}  The  brightness  on  the  upper  part  of  the  proscenium  due  to  the  close 
proximity  of  the  two  front  luminaires  became  somewhat  disturbing.  This 
brightness  was  1.8  foot-lamberts. 

Objection  No.  1  might  be  relieved  somewhat  by  a  shadow-box  ar- 
rangement, and  objection  No.  2  by  relocating  the  front  luminaires. 

A  second  test  was  made  with  this 
same  indirect  system  providing  0.25  foot- 
candle  in  the  auditorium.  Objection 
No.  2  was  now  overcome  but  the  light 
spilled  on  the  screen  was  still  objection- 
able, as  it  measured  0.18  foot-lambert. 

Test  No.  6 — Luminous  Elements  (Win- 
dows) at  Upper  Side  Walls. — Because  the 
junction  of  ceiling  and  side  walls  is  farthest 
from  the  line  of  vision  when  viewing  the 
picture,  it  was  thought  that  it  would  be 
found  that  highest  brightnesses  would  be 
acceptable  here.  Tests  with  the  lighted 
windows  substantiated  this  fact.  How- 
ever, it  was  apparent  that  there  was  a 
decided  distraction  from  a  concentration 
of  illumination  at  these  positions  which 
conflicted  definitely  with  the  screen.  As 
a  result,  even  though  higher  brightnesses 
did  not  appear  glaring  here,  it  was  con- 
cluded that  a  low  order  of  brightness,  especially  at  the  front  of  the 
auditorium,  no  higher  than  a  brightness  of  3  foot-lamberts  should  be 
used,  which  is  in  accord  with  the  findings  of  Jones.4  However,  this 
brightness  may  be  gradually  increased  toward  the  rear  of  the  audi- 
torium as  the  subtended  angle  at  the  eye  of  screen  and  light-source 
becomes  greater.  Many  theaters  have  as  their  sole  source  of 
illumination  shaded  side- wall  brackets.  With  a  good  combination 
having  two  10- watt  amber  lamps  the  brightest  point  measured  21 
foot-lamberts.  Two  15- watt  lamps  gave  a  measurement  of  65  foot- 
lamberts.  The  light  oak  background  measured  8  foot-lamberts  and 


FIG.  4.  The  downlight- 
ing  system  used  in  Test 
No.  7,  which  affords  one  of 
the  best  methods  now 
available  of  directing  and 
controlling  light  for  audi- 
torium purposes. 


208  F.  M.  FALGE  AND  W.  D.  RIDDLE          [j.  s.  M.  P.  E. 

with  a  white  31  foot-lamberts.  These  values  are  all  above  the  one 
for  comfort,  indicating  that  this  method  of  lighting  is  in  general 
undesirable. 

Test  No.  7 — Downlighting. — (Fig.  3)  As  previously  pointed  out, 
the  downlighting  system  in  this  auditorium  was  of  the  objective-lens 
type,  and  it  was  so  directed  as  to  direct  light  forward  of  the  vertical 
(Fig.  4) .  Brightness  at  the  ceiling  openings  was  low. 

With  downlighting  alone,  illumination  could  be  raised  to  a  point 
providing  0.08  foot-candle  in  the  auditorium  before  the  brightness 
on  the  heads  of  persons  in  the  beams  directly  beneath  the  units  be- 
came disturbing.  At  other  locations  values  of  0.2  to  0.25  foot-candle 
were  satisfactory.  At  no  point  did  illumination  fall  upon  the  screen 
so  as  to  be  discernible,  and  ceiling  or  side- walls  were  not  lighted  so  that 
there  was  no  objectionable  brightness  here. 

Test  No.  8 — Downlighting  and  Indirect  Lighting. — By  adding  0.04 
foot-candle  of  indirect  lighting  the  brightness  of  the  direct  downward 
light  was  relieved  and  the  system  was  then  quite  comfortable. 

An  additional  test  was  made  by  adding  0.25  foot-candle  of  down- 
lighting  to  0.25  foot-candle  of  indirect  lighting  to  provide  the  maxi- 
mum of  0.5  foot-candle,  and  this  system  was  very  comfortable.  Pic- 
ture quality  was,  however,  impaired  by  spilled  light  on  the  screen 
from  the  indirect  lighting. 

Conclusions  as  a  result  of  these  tests  and  others  cited  were  as  fol- 
lows for  the  given  auditorium  conditions : 

(1)  A  picture  viewed  without  any  auditorium  illumination  is  definitely  un- 
comfortable. 

(2)  A  minimum  illumination  of  0.05  to  0.1  foot-candle  is  required  to  soften 
picture  contrast  with  background.    Variation  of  picture  size  and  brightness,  and 
auditorium  size  would  doubtless  influence  this  figure.    Jones4  found  that  an  il- 
lumination of  0.1  foot -candle  at  the  front  and  0.2  at  the  rear  of  the  auditorium 
is  desirable,  and  this  is  checked  by  the  report  before  the  International  Com- 
mission of  Illumination.9 

(5)  A  maximum  illumination  of  0.5  foot-candle  is  permissible  from  the  stand- 
point only  of  distraction  caused  by  the  movements  of  other  persons. 

(4)  Some  illumination — x/25  to  Vso  of  screen  brightness — is  desirable  for  the 
immediate  screen  surroundings.     Higher  relative  brightnesses  were  satisfactory 
at  greater  angles  from  the  screen.    At  the  outer  edges  of  the  proscenium,  values 
of  2  to  3  foot-lamberts  were  permissible,  and  approaching  the  rear  of  the  audi- 
torium considerably  higher  values  are  acceptable,  depending  upon  their  height 
above  eye  level. 

(5)  Indirect  lighting  has  many  advantages  for  auditorium  lighting  because 
the  light  is  spread  so  as  to  be  low  in  brightness  and  because  an  illuminated  ceiling 
adds  to  comfort.    But  it  is  desirable  to  control  the  light  at  the  front  of  the  audi- 


Feb.,  1939]  LIGHTING  THEATER  AUDITORIUMS  209 

torium  so  that  the  brightness  at  points  near  the  junction  of  ceiling  and  proscenium 
is  below  2  to  3  foot-lamberts  and  the  screen  brightness  contributed  by  spilled 
light  is  below  0.05  foot-lambert. 

(6)  Concentration  of  light  at  the  front  side  walls  is  distracting.    Such  sources 
should  be  kept  below  3  foot-lamberts.    Side- wall  brackets  are  in  general  too  bright 
for  comfortable  vision. 

(7)  Downlighting  by  controlled  beams  of  light  needs  to  be  supplemented  by 
some  indirect  lighting.    The  combination  of  the  two  systems  affords  the  best  see- 
ing conditions  of  any  investigated,  minimizing,  as  it  does,  bright  conflicting  sources 
and  spilled  light  on  screen. 


REFERENCES 

1  LUCKIESH,  M.,  AND  Moss,  F.  K.,  "The  Motion  Picture  Screen  as  a  Lighting 
Problem,"  /.  Soc.  Mot.  Pict.  Eng.,  XXVI  (May,  1936),  p.  578 

2  FALGE,  F.  M.,  AND  WEITZ,  C.  E.:   "Theatre  Lighting,"  Bulletin,  Nela  Park 
Engineering  Department,  General  Electric  Company. 

3  CAMBRIA,  F.,  AND  FALGE,  F.  M.:    "Theatre  Lighting,  Its  Tragedies,  Its 
Virtues,"  Trans.  Ilium.  Eng.  Soc.,  XXIV  (Nov.,  1929),  p.  890. 

4  JONES,  L.  A.:    "The  Interior  Illumination  of  the  Motion  Picture  Theater," 
Trans.  Soc.  Mot.  Pict.  Eng.,  IV  (1920),  No.  10. 

6  LUCKIESH,  M.,  AND  TAYLOR,  A.  H.:    "A  Brightness  Meter,"  J.  Opt.  Soc. 
Amer.,  27  (March,  1937),  p.  132. 

6  SCHLANGER,  B.:    "A  Method  of  Enlarging  the  Visual  Field  of  the  Motion 
Picture,"  /.  Soc.  Mot.  Pict.  Eng.,  XXX  (May,  1938),  p.  503. 

7  O'BRIEN,  B.,  AND  TUTTLE,  C.  M.:   "An  Experimental  Investigation  of  Pro- 
jection Screen  Brightness,"  /.  Soc.  Mot.  Pict.  Eng.,  XXVI  (May,  1936),  p.  505. 

8  WOLF,  S.  K.:    "An  Analysis  of  Theater  and  Screen  Illumination  Data," 
J.  Soc.  Mot.  Pic.  Eng.,  XXVI  (May,  1936),  p.  532. 

9  Report  of  the   (Japanese)   Secretariat  Committee,  Proc.  Internal.    Comm. 
Ilium.,  Cambridge  (England),  8th  Session  (Sept.,  1931). 

10  Report  of  the  Theater  Lighting  Committee,  /.  Soc.  Mot.  Pict.  Eng.,  XIV 
(Feb.,  1931),  p.  239. 

DISCUSSION 

MR.  CRABTREE  :     The  spots  on  the  ceiling  were  somewhat  annoying  to  me. 

MR.  FALGE:  The  picture  does  seem  to  give  that  impression.  In  the  actual 
condition  you  would  not  see  the  spots  or  would  not  be  conscious  of  them 
because  they  are  well  located.  When  there  is  a  vertical  cut-off  and  the  eyes  are 
not  in  the  beam  of  the  light  the  result  would  not  be  glaring.  The  illustrations 
do  not  give  the  proper  impression. 

MR.  SCHLANGER:  It  is  unfortunate  that  so  careful  a  study  was  not  made  in  a 
room  having  more  suitable  lighting  arrangements.  The  direct-indirect  type  of 
lighting  fixtures  suspended  from  the  ceiling  are  glare  spots  in  the  patrons'  field 
of  vision.  The  glass  panel  on  the  underside  of  the  suspended  fixtures  in  the  direct 
lighting  portion  is  the  most  disturbing  element;  however,  the  indirect  light  com- 
ing from  the  tops  of  these  isolated  and  suspended  fixtures  is  also  objectionable. 


210  F.  M.  FALGE  AND  W.  D.  RIDDLE          [j.  s.  M.  P.  E. 

Apparently  the  authors  used  these  fixtures  merely  to  help  determine  illumina- 
tion levels,  and  they  were  not  intended  for  recommended  use.  There  are  systems 
of  indirect  lighting  that  are  more  adaptable  to  this  use;  however,  the  use  of 
secondary  illumination  of  any  kind  during  a  screen  presentation  is  debatable. 
The  elimination  of  glare  spots  and  contrasting  levels  of  illumination  during  the 
screen  presentation  are  exceedingly  important.  I  have  found  that  the  general 
level  of  illumination  can  be  increased  to  a  surprising  degree  when  the  surfaces 
presented  to  the  eyes  are  uniformly  illuminated  and  devoid  of  all  patterns  caused 
by  highlights,  shadows,  or  contrasting  decorations. 

Secondary  illumination  of  the  auditorium  surfaces  during  the  screen  presenta- 
tion is  most  costly  if  carried  out  properly  because  the  surfaces  are  usually  dark 
enough  to  absorb  screen  light,  thereby  diminishing  the  efficiency  of  other  forms 
of  lighting.  It  seems  that  for  greatest  economy  and  simplicity,  light  surfaces 
reflecting  the  screen  light  would  be  efficient  provided  the  reflections  were  con- 
trolled by  the  texture  or  angularity  of  the  surfaces.  Mr.  Falge,  what  are  the  in- 
tensity and  color  of  the  walls  of  the  auditorium  where  the  tests  were  made? 

MR.  FALGE:  Light  ceiling  and  side  walls,  and  a  very  light  oak  panelling  on 
the  lower  side  wall.  The  color:  light  oak  finish,  a  sort  of  tan. 

MR.  SCHLANGER:  I  have  had  experience  with  theaters  where  the  walls  were 
pure  white  as  well  as  where  they  were  darker.  Surprisingly  enough,  in  some  of 
the  theaters  where  the  walls  were  pure  white  the  effect  was  not  as  objectionable 
as  might  be  expected.  If  all  the  screen  light  reflected  from  surfaces  is  returned 
to  the  viewers'  eyes  an  annoying  flicker  would  result.  To  avoid  this  the  texture 
or  angularity  of  the  surfaces  could  be  such  as  to  reflect  a  greater  part  of  the  light 
so  that  it  falls  upon  the  heads  and  sides  of  the  viewers,  leaving  a  small  percentage 
of  the  light  to  reflect  to  the  eyes.  This  would  provide  desirable  diffuse  lighting 
of  the  audience,  and  would  also  light  up  the  wall  and  ceiling  surfaces  enough  to 
avoid  sharp  contrast  between  the  picture  and  the  auditorium  surface  light. 

A  certain  amount  of  secondary  light  can  be  used  in  addition  to  reflected  screen 
light.  Such  secondary  light-sources  must  be  located  on  wall  or  ceiling  surfaces 
that  do  not  fall  within  the  field  of  vision  of  the  audience.  The  type  of  secondary 
lighting  so  employed  would  be  optional,  only  efficiency  and  appearance  being  the 
important  considerations.  Direct  lighting  sources  more  efficient  than  the  indirect 
can  be  successfully  used.  The  secondary  lighting  serves  also  as  emergency  light- 
ing in  the  event  of  a  break  in  the  screen  light. 

Motion  picture  theater  auditoriums  have  been  and  are  being  designed  ac 
cording  to  Spanish,  Aztec,  or  Modernistic  inspirations.  Such  forms  are  unsuited 
to  the  purpose  of  the  motion  picture  theater  for  two  basic  reasons.  First,  such 
decorations,  which  necessarily  consist  of  forms  creating  shadows  and  highlights 
and  differences  in  color  and  intensity  in  paint  design,  create  the  disturbing  con- 
trasts referred  to  before;  second,  such  decoration  quite  often  becomes  an 
unsuitable  setting  for  the  subjects  of  the  films. 

It  is  possible  to  create  an  abstract  form  of  decoration,  using  as  inspiration  such 
textures  and  angularities  of  surfaces  as  will  properly  react  to  screen  light. 

It  should  be  pointed  out  that  the  screen  size  in  relation  to  the  auditorium  size 
referred  to  in  these  tests  was  rather  small,  and  therefore  the  factor  of  screen  light 
was  not  fully  accounted  for. 

MR.  FRANK:     Has  any  study  been  given  to  the  color  of  light  most  desirable  in 


Feb.,  1939]  LIGHTING  THEATER  AUDITORIUMS  211 

a  theater,  or  the  use  of  fluorescent  lights  in  auditoriums?  Also,  have  any  studies 
been  made  as  to  desirable  aisle  lighting? 

MR.  FALGE:  With  regard  to  Mr.  Schlanger's  questions,  the  under  surface 
of  the  suspended  luminaires  was  a  source  of  annoyance  and  should  have  been 
omitted.  The  system  was  not  the  best  one  for  indirect  lighting,  and  much  im- 
provement could  be  made  on  it.  As  to  the  current  consumption,  there  is  consider- 
able misunderstanding  on  the  part  of  many  theater  operators  as  to  what  consti- 
tutes current  consumption.  I  have  visited  theaters  that  had  side-wall  brackets 
with  one  bulb  in  each  bracket  and  a  total  of  about  100  watts  for  the  auditorium 
lighting,  and  have  been  told,  "I  want  my  lighting  improved,  but  I  do  not  want 
to  use  any  more  current."  Theater  operators  are  inclined  to  think  of  their 
lighting  bill  as  their  projection  bill.  They  forget  that  perhaps  the  greater  por- 
tion of  the  bill  goes  into  projection  and  that  the  few  lamps  mentioned  above 
amount  to  very  little. 

With  systems  such  as  downlighting,  and  with  better  application  of  color  and 
more  efficient  light-sources  and  equipment,  it  is  feasible  to  light  an  auditorium 
economically  to  the  low  levels  about  which  we  are  talking.  It  is  the  higher  levels 
needed  for  intermission  lighting  that  require  more  current,  and  yet  the  lights  are 
used  for  such  a  short  space  of  time  that  they  are  not  excessively  costly. 

Mr.  Schlanger's  suggestion  of  reflecting  the  screen  light  from  the  side  walls  is 
worth  considerable  study.  The  light  can  be  utilized  as  he  suggests,  and  it  would 
be  interesting  to  see  whether  it  could  be  done  practically. 

As  to  the  suggestion  that  the  theater  auditorium  should  be  entirely  simple, 
with  no  special  decorations  at  all,  there  is  some  question  on  that  score.  I  listed 
other  factors  that  are  still  of  importance  to  the  theater  manager  and  owner. 
The  patrons  react  to  lighted  interiors  in  other  ways,  even  though  our  study  was 
devoted  primarily  to  seeing. 

Referring  to  Mr.  Frank's  question,  we  could  not  go  into  the  study  of  color 
this  time.  It  is  a  difficult  study  to  make,  the  conditions  are  difficult  to  stabilize, 
and  reactions  are  very  difficult  to  get.  That  is  a  problem  for  the  future,  and  an 
important  one.  However,  we  have  studied  the  color  of  light  as  it  relates  to  the 
efficiency  of  systems.  Fluorescent  lighting  has  great  possibilities  for  theater 
interiors  because  of  the  extremely  high  efficiency  that  is  attainable.  With  the 
green,  for  instance,  we  get  60  lumens  per  watt,  whereas  with  the  regular  green 
lamp  only  one  lumen  per  watt.  We  can  get  efficiencies  up  to  100  times  as  much 
as  with  the  usual  colored  light-sources. 

Aisle  lighting  was  included  in  our  study.  With  little  lighting  in  the  auditorium, 
aisle  lighting  does  help  to  identify  the  aisles,  which  seems  to  be  its  primary  aim. 
However,  with  some  of  these  other  systems,  well  planned  downlighting  is 
used  over  the  aisles,  and  in  that  case  the  aisle  lighting  is  brought  out  very  ex- 
cellently through  that  means. 

MR.  CRABTREE  :  I  think  the  keynote  of  this  paper  and  discussion  is  to  avoid 
distraction.  If  we  could  only  get  that  idea  over  to  the  architects  and  publicize 
it  to  the  same  extent  that  the  report  of  the  Projection  Practice  Committee  is  being 
publicized,  we  would  really  be  getting  somewhere.  In  Rochester  one  of  the 
principal  theaters  has  just  been  redecorated,  and  apparently  the  management  is 
highly  pleased  with  it;  but  when  you  go  downstairs  and  sit  under  the  balcony 
there  are  glaring  lights  almost  as  intense  as  those  from  the  side  walls  blazing  into 


212  F.  M.  FALGE  AND  W.  D.  RIDDLE 

your  eyes,  so  that  enjoyment  of  the  picture  is  impossible.  It  is  quite  apparent 
therefore  that  some  architects  are  absolutely  ignorant  of  the  fundamentals  in- 
volved. 

As  I  have  said  before,  to  me  the  ideal  condition  is  a  completely  darkened  room. 
Experiments  have  shown,  however,  that  the  general  level  can  be  raised  tre- 
mendously before  the  visibility  of  the  picture  is  impaired,  and  under  those 
conditions  I  do  not  think  you  need  aisle  lighting.  In  the  rear-projection  theaters, 
the  general  level  is  tremendously  high,  but  the  contrast  of  the  picture  is  pleasing 
and  adequate. 

MR.  FALGE:  I  think  Mr.  Crabtree's  comments  are  absolutely  right.  That 
is  the  keynote  of  everything  we  have  found.  Eliminate  distraction  and  you  are 
much  nearer  a  condition  of  comfort. 

MR.  WOLF:     It  is  impracticable,  however,  to  have  complete  darkness. 

MR.  FALGE  :     I  think  it  is. 

MR.  CARLSON  :  I  agree  with  Mr.  Crabtree  that  the  elimination  of  distracting 
influences  is  certainly  necessary.  Another  way  of  expressing  it  would  be  that  we 
are  interested  primarily  in  maintaining  low  differences  in  brightness.  In  other 
words,  as  in  rear-projection  theaters,  a  relatively  high  general  level  of  illumination 
may  be  not  only  acceptable  but  actually  pleasing,  so  long  as  there  are  no  sources 
of  excessive  brightness  in  the  field  of  view. 

MR.  SCHLANGER:  I  have  tried  out  some  fairly  successful  indirect  systems  in 
theaters,  for  use  during  the  screen  performances,  that  have  not  proved  distracting. 
Unfortunately,  these  systems  are  used  only  during  intermissions,  because  the 
theater  operators  have  found  them  expensive  for  continuous  operation.  For 
best  results,  indirect  lighting  must  be  continuous  and  uninterrupted.  Indirect 
lighting  is  inefficient  because  of  the  amount  of  light  that  must  necessarily  be 
trapped.  For  these  reasons,  it  is  too  costly  for  daily  operating  periods  of  ap- 
proximately eleven  hours. 

The  special  holiday  or  other  decorative  lighting  effects  to  which  Mr.  Falge 
refers  are  desirable,  and  can  be  incorporated  in  a  built-in  manner  in  any  lighting 
scheme,  but  such  lighting  is  intended  purely  for  intermissions  and  would  be 
highly  distracting  during  screen  performances. 

With  regard  to  Mr.  Crabtree's  statement  about  the  rear-projection  theaters, 
I  have  found  that  regular  front-projection  theaters  can  be  illuminated  to  levels 
as  high  as  or  higher  than  the  levels  found  in  the  rear-projection  theaters. 


REVISED  STANDARD  ELECTRICAL  CHARACTERISTICS 
FOR  TWO-WAY  REPRODUCING  SYSTEMS 
IN  THEATERS* 


RESEARCH  COUNCIL,  ACADEMY  OF  MOTION  PICTURE  ARTS 
AND  SCIENCES** 


(As  a  result  of  additional  tests  and  consideration  of  field  operating  conditions, 
these  Revised  Specifications  are  recommended  to  supersede  the  original  Specifications 
of  March  31,  1937,  and  the  subsequent  Specifications  of  June  8,  1937.) 

Systems  to  Which  These  Specifications  Apply:  The  two-way  repro- 
ducing systems  for  which  these  characteristics  are  recommended,  are : 

Type  7— ERPI  Mirrophonic  Systems  M-101,  M-l,  M-2,  and  M-3 
using  594-A  loud-speaker  units  (metal  diaphragm)  and  TA-4181-A 
low-frequency  units,  M-4  using  555  loud-speaker  units  (metal  dia- 
phragm) and  TA-4181-A  low-frequency  units,  and  M-5  using  555 
loud-speaker  units  (metal  diaphragm)  and  T A -4194  low-frequency 
units. 

Type  II — RCA  system  using  M. I. -143 5  (metal  diaphragm)  and 
MJ.-1432-A  low-frequency  mechanisms. 

Type  III — Lansing  equipped  system  using  284  or  285  (metal 
diaphragm)  and  15X  low-frequency  mechanisms. 

Type  IV—  RCA  system  using  M.I.-1428-B  or  M.I.-1443  (non- 
metallic  diaphragm)  and  MJ.-1432-A  or  M.I. -1444  low-frequency 
mechanisms. 

Measurement  Point:  These  characteristics  are  valid  for  measure- 
ments made  at  the  output  of  the  power  amplifier,  including  the 
low-pass  filter,  with  a  resistance  equivalent  to  the  speaker  load, 
using  the  Academy  Research  Council  Standard  Multi-Frequency 
Test  Reel  (corrected),!  and  are  subject  to  modifications  to  fit  special 
acoustic  conditions  which  exist  in  many  theaters,  due  to  the  fact 
that  the  reverberation  time  or  other  acoustic  characteristics  are  not 
optimum. 

*  Reprinted  from  Technical  Bulletin,  Research  Council,  Academy  of  Motion 
Picture  Arts  &  Sciences,  October  10,  1938. 
**  Hollywood,  Calif. 

f  The  correction  factors  indicate  the  deviation  from  constant  percentage 
modulation  for  each  frequency. 

213 


214  STANDARD  ELECTRICAL  CHARACTERISTICS   [J.  S.  M.  P.  E. 

The  above  Academy  Research  Council  Test  Reels  have  been  com- 
pared to  both  the  Altec  ED -20  (corrected),*  and  the  RCA  test  film 
(Catalogue  No.  26571),  and  all  these  reels  should  give  approximately 
the  same  characteristic  on  any  one  equipment. 

Extensive  field  tests  indicate  that  equipment  set  to  these  Standard 
Electrical  Characteristics  will  give  optimum  reproduction  of  current 
studio  recordings  under  all  conditions. 

It  has  also  been  found  that  the  calibration  of  individual  frequency 
reels  in  use  in  the  field  varies  in  some  instances.  In  case  results 
obtained  from  any  of  these  reels  are  obviously  in  error,  the  calibra- 
tion of  the  test  reel  used  in  making  the  run  should  be  checked. 

Acoustic  Correction:  Whenever  such  conditions  exist  that  the 
particular  characteristic  recommended  does  not  give  satisfactory 
results  (after  the  calibration  of  the  reel  used  for  the  frequency  run 
has  been  checked),  it  is  recommended  that  the  acoustic  characteristics 
of  the  auditorium  be  corrected. 

Mechanism  Adjustment:  With  the  available  equipment  as  specified, 
operating  with  the  Standard  Electrical  Characteristic,  it  is  necessary 
in  some  instances  that  the  sensitivity  of  the  high-  and  low-frequency 
band  be  relatively  adjusted  to  obtain  a  flat  acoustic  response  on 
both  sides  of  the  cross-over.  This  adjustment  usually  takes  the  form 
of  attenuating  the  high-frequency  band,  the  amount  of  this  attenua- 
tion depending  upon  the  relative  efficiency  of  both  low-  and  high- 
frequency  units  and  the  specific  properties  of  the  auditorium  involved. 

Typical  values  are  as  follows : 

Type  I—  ERPI  Mirrophonic  systems,  M-101,  M-l,  M-2,  M-3,  and 
M-5,  attenuate  the  high-frequency  band  2  to  4  db. 

Type  I — ERPI  Mirrophonic  system,  M-4,  attenuates  the  high- 
frequency  band  0  to  2  db. 

Type  //—RCA  systems,  M. I. -1435  and  M.I.-1432-A,  attenuate 
the  high-frequency  band  0  to  2  db. 

Type  III — Lansing  equipped  systems  attenuate  the  high-frequency 
band  0  to  2  db. 

Type  IV—  RCA  systems,  M.I.-1428-B,  M.I.-1432-A,  M.L-1443, 
and  M.I.-1444,  attenuate  the  low-frequency  band  0  to  2  db. 

Note:  It  should  be  remembered  that  the  type  and  condition  of 
screen  used  in  the  theater  will  in  a  measure  affect  the  high-frequency 
response  of  the  reproducing  system. 

Tolerance:  A  tolerance  of  =±=1  db  up  to  3000  cycles,  increasing 
progressively  with  frequency  to  =±=2  db  at  7000  cycles,  is  the  maxi- 
mum permitted  for  the  following  gain-frequency  measurements. 


Feb.,  1939]  STANDARD  ELECTRICAL  CHARACTERISTICS 


215 


S     S    3       Si 


I       I    I   I  I  HII 


FIG.  1.  Revised  Standard  Electrical  Characteristic  for  two-way  reproduc- 
ing systems  using  metal  diaphragms;  Types  I  (M-101,  M-l,  M-2  Systems), 
II,  and  III.  For  optimum  results  with  current  studio  sound  recordings  Type 
I,  II,  and  III  Systems  equipped  with  metal  diaphragm  speakers  should  be 
adjusted  to  this  Revised  Standard  Electrical  Characteristic. 


S      S    3       Si 


i       !     S    i  S  Mil 


FIG.  2.  Standard  Electrical  Characteristic  for  two-way  reproducing  systems 
using  metal  diaphragms ;  Type  I  (M-3  Systems) .  This  characteristic  for  Type 
I  equipments  (M-3  Systems)  has  not  been  changed,  and  remains  as  specified 
in  the  original  publication  of  March  31,  1937,  and  the  subsequent  publication 
of  June  8,  1937. 

Electrical  Runs,  Measured  at  the  Output  of  the  Power  Amplifier  with  a  Resistance 
Equivalent  to  the  Speaker  Load  Using  the  Academy  Research  Council  Standard 
Multi-Frequency  Test  Reel  (Corrected),  Altec  Test  Film  (ED-20,  Corrected),  or 

RCA  Test  Film  (Catalogue  No.  26571) 

The  tolerances  of  ±1  db.  up  to  3000  cycles,  increasing  progressively  with  fre- 
quency to  a  maximum  of  =*=2  db.  at  7000  cycles,  should  be  rigidly  maintained  in 
adjusting  equipment  to  these  specifications. 

Electrical  Runs,  Measured  at  the  Output  of  the  Power  Amplifier  with  a 
Resistance  Equivalent  to  the  Speaker  Load  Using  the  Academy  Research 
Council  Standard  Multi-Frequency  Test  Reel  (Corrected),  Altec  Test  Film 
(ED-20,  Corrected) ,  or  RCA  Test  Film  (Catalogue  No.  26571) 

The  tolerances  of  ±1  db.  up  to  3000  cycles,  increasing  progressively  with 
frequency  to  a  maximum  of  =*=2  db.  at  7000  cycles,  should  be  rigidly  main- 
tained in  adjusting  equipment  to  these  specifications. 


216 


STANDARD  ELECTRICAL  CHARACTERISTICS 


II  ill 


S       8  I 


i  i  i  nil 


FIG.  3.  Revised  Standard  Electrical  Characteristic  for  two-way  repro- 
ducing systems  using  metal  diaphragms;  Type  I  (M-4,  M-5  Systems).  For 
optimum  results  with  current  studio  sound  recordings,  Type  I  systems 
equipped  with  metal  diaphragm  speakers  should  be  adjusted  to  this  revised 
Standard  Electrical  Characteristic. 


I    i  i  tiiil 

IN  srcn»  PCX  tccow 


FIG.  4.  Revised  Standard  Electrical  Characteristic  for  two-way  repro- 
ducing systems  using  RCA  non-metallic  diaphragms;  Type  IV.  For 
optimum  results  with  current  studio  sound  recordings,  those  two-way  repro- 
ducing systems  equipped  with  non-metallic  diaphragm  speakers  (Type  IV) 
should  be  adjusted  to  this  revised  Standard  Electrical  Characteristic. 

Electrical  Runs,  Measured  at  the  Output  of  the  Power  Amplifier  with  a  Resistance 
Equivalent  to  the  Speaker  Load  Using  the  Academy  Research  Council  Standard 
Multi-Frequency  Test  Reel  (Corrected),  Altec  Test  Film  (ED-20,  Corrected),  or 

RCA  Test  Film  (Catalogue  No.  26571) 

The  tolerances  of  ±1  db.  up  to  3000  cycles,  increasing  progressively  with 
frequency  to  a  maximum  of  ±2  db.  at  7000  cycles,  should  be  rigidly  maintained 
in  adjusting  equipment  to  these  specifications. 

Electrical  Runs,  Measured  at  the  Output  of  the  Power  Amplifier  with  a 
Resistance  Equivalent  to  the  Speaker  Load  Using  the  Academy  Research 
Council  Standard  Multi-Frequency  Test  Reel  (Corrected),  Altec  Test  Film 
(ED-20,  Corrected),  or  RCA  Test  Film  (Catalogue  No.  26571) 

The  tolerances  of  ±1  db.  up  to  3000  cycles,  increasing  progressively  with 
frequency  to  a  maximum  of  ±2  db  at  7000  cycles,  should  be  rigidly  main- 
tained in  adjusting  equipment  to  these  specifications. 


ORGANIZATION  OF  THE  WORK  OF  THE  PAPERS 
COMMITTEE* 


Since  a  new  committee  Chairman  will  be  appointed  for  the  year 
1939,  the  present  Chairman  has  considered  it  desirable  to  place  on 
record,  for  the  guidance  of  future  chairmen,  a  somewhat  detailed 
account  of  the  procedures  involved  in  preparing  the  papers  programs 
for  our  Semi- Annual  Conventions,  as  follows. 

Committee  Personnel. — Experience  has  shown  that  an  increasing 
number  of  the  papers  in  our  JOURNAL  are  being  written  by  technicians 
in  the  studios  and  laboratories  on  the  West  Coast.  In  February,  1937, 
at  the  suggestion  of  H.  G.  Tasker,  Past- President  of  the  Society,  a 
sub-committee  of  the  Papers  Committee  was  formed  on  the  West 
Coast  with  W.  A.  Mueller  as  Chairman  and  L.  A.  Aicholtz  as  Secre- 
tary. Each  of  the  major  studios  and  laboratories  was  represented. 
This  plan  centralized  the  work  of  paper  solicitation  on  the  West 
Coast  and  has  proved  a  very  practicable  arrangement.  It  is  strongly 
urged,  therefore,  that  future  chairmen  adopt  a  similar  plan,  so  that 
there  will  always  be  an  active  and  representative  sub-committee  on 
the  West  Coast. 

Other  members  of  the  general  Papers  Committee  should  be  chosen 
from  the  leading  industrial  firms  in  the  East  and  Middle  West  in  order 
that  a  direct  relationship  will  be  established  with  the  principal  labora- 
tories and  branches  of  the  industry.  Members  should  also  be  selected 
in  Europe  in  those  countries  where  cinematographic  research  programs 
are  known  to  be  in  progress.  Since  the  work  of  the  Committee  must 
be  carried  on  largely  by  correspondence,  the  personnel  should  not  be 
too  large. 

ORGANIZATION  OF  COMMITTEE  WORK 

Details  of  the  work  may  be  classified  as  follows  in  the  order  of  time 
before,  during,  and  after  a  meeting : 

(1)  Copy  for  Journal  Notice  of  the  Meeting. — The  copy  for  a  request 
for  papers  should  be  published  in  each  issue  of  the  JOURNAL  for  four 
months  before  the  meeting.  This  copy  should  be  prepared  about  one 
month  after  the  close  of  one  meeting.  A  typical  notice  is  the  follow- 

*  Received  December  28,  1938, 

217 


218  WORK  OF  PAPERS  COMMITTEE  [j.  s.  M.  P.  E. 

ing  one  used  for  the  1938  Fall  Meeting  (Detroit,  Mich.,  Oct.  31-Nov. 
2,  1938).  This  notice  was  printed  on  the  inside  cover  of  the  JOURNAL 
for  June,  July,  August,  and  September,  1938. 

PAPERS  FOR  THE  FALL  CONVENTION 

Manuscripts  of  papers  received  by  September  1st  will  be  given  immediate 
consideration  by  the  Papers  Committee  and  the  Board  of  Editors,  and  the  best 
will  be  selected  and  given  preferred  positions  on  the  program  of  the  Convention, 
with  ample  time  for  presentation  and  discussion,  or  about  30  minutes  to  one 
hour. 

Titles  and  abstracts  of  all  papers  must  be  received  by  September  15th  to  be  con- 
sidered for  listing  on  the  Preliminary  Program. 

Two  complete  copies  of  each  manuscript  must  be  sent  to  the  Chairman  of  the 
Papers  Committee  by  October  1st,  in  order  that  the  paper  be  listed  on  the  Final 
Program  for  presentation.  Manuscripts  arriving  after  October  1st  may,  at  the 
discretion  of  the  Papers  Committee,  be  scheduled  on  the  Program  to  be  read  by 
title  or  substituted  for  other  papers  in  the  event  of  cancellations. 

(2)  Preliminary  Letters. — (a)  About  three  months  before  the  meet- 
ing, all  members  of  the  Committee  should  be  circularized  by  letter, 
giving  dates  of  meeting,  closing  dates  for  titles,  abstracts,  and  manu- 
scripts.   Previous  Committee  correspondence  should  be  studied  and 
a  summary  given  in  each  letter  of  any  papers  which  have  been  held 
over  from  previous  meetings. 

(b)  A  prospect  file  should  be  kept  for  each  meeting  and  every  pros- 
pect should  be  sent  a  letter  of  inquiry  on  the  status  of  a  paper  for  the 
next  meeting. 

(c)  The  condition  of  the  industry  should  be  analyzed  and  letters 
sent  to  possible  authors  of  papers  on  subjects  of  current  interest. 

(d)  A  letter  should  be  sent  to  each  of  the  Vice-Presidents  of  the 
Society  inquiring  as  to  the  possibility  of  reports  from  the  Committees 
under  their  supervision. 

(3)  Preparation  of  Authors'  Form. — This  form  may  be  mimeo- 
graphed.   Copies  should  be  sent  to  each  member  of  the  Committee 
about  two  weeks  after  the  first  letter  with  a  letter  of  reminder.    A 
typical  form  is  attached  to  this  report.    As  favorable  replies  are  re- 
ceived from  the  letters  sent  to  prospective  authors,  an  Authors'  Form 
should  be  sent  out  accompanied  by  a  reprint  of  the  Regulations  of 
the  SMPE  Related  to  the  Preparation  of  Papers  for  Presentation  and 
Publication  (J.  Soc.  Mot.  Pict.  Eng.,  31,  215,  Aug.,  1938).    Each  com- 
mittee member  and  author  should  be  informed  of  the  necessity  that 
abstracts  be  sent  in  by  the  date  specified  and  that  two  copies  of  each 
manuscript  must  be  delivered  by  the  date  indicated. 


Feb.,  1939]  WORK  OF  PAPERS  COMMITTEE  219 

(4)  Abstracts  of  the  Papers  to  Journal  Editor. — It  is  usually  neces- 
sary to  revise  some  of  the  abstracts  as  received  from  the  authors,  and 
all  abstracts  should  preferably  be  retyped  to  give  a  desirable  uni- 
formity of  copy  for  the  printer.    Some  abstracts  are  too  long;   some 
are  written  poorly  as  to  sentence  construction;  and  occasionally  one 
is  received  that  is  advertising  copy  rather  than  an  informative,  con- 
cise statement  of  the  author's  paper.    It  is  important,  therefore,  that 
every  abstract  be  read  carefully  before  release  for  printing  in  the 
JOURNAL.    Abstract  copy  should  be  sent  to  the  editor  in  time  to  ap- 
pear in  the  JOURNAL  issued  prior  to  the  meeting. 

Galley  proof  of  abstracts  should  also  be  read  to  pick  up  printing 
errors,  especially  with  regard  to  names  of  authors  and  their  company 
affiliations.  Extra  sets  of  galleys  of  the  abstracts  should  be  supplied 
the  Chairman  of  the  Publicity  Committee  for  the  use  of  the  trade 
publications.  These  should  not  be  released,  however,  until  publica- 
tion has  been  made  in  the  JOURNAL.  Arrangements  for  these  details 
can  be  made  with  the  Editorial  Office. 

(5)  Preparation  of  Preliminary  Program. — The  preparation  of  the 
program  requires  a  careful  study  of  the  abstract  of  each  paper  before 
a  suitable  arrangement  can  be  made  of  the  papers  under  various 
special  headings.     These  "sessions"  or  "symposiums"  have  proved 
an  effective  and  popular  scheme  for  concentrating  the  interest  and 
stimulating  discussion  at  our  semi-annual  meetings.     It  is  recom- 
mended that  this  plan  be  continued. 

The  arrangement  of  the  papers  should  also  be  planned  with  regard 
to  the  work  of  the  Publicity  Committee.  For  example,  papers  of 
news  value  should  preferably  be  scheduled  on  the  first  and  second 
days  of  the  meeting  and  distributed  to  bring  at  least  one  such  paper 
in  each  morning  and  afternoon  session. 

When  meetings  are  held  in  the  East,  it  is  recommended  that  the 
technical  paper  sessions  be  restricted  to  the  daytime  and  held  in  the 
evening  only  if  the  subject  matter  is  very  outstanding  and  is  accom- 
panied by  a  demonstration.  Examples  of  such  evening  sessions  are: 
The  paper  on  "Color  Photography"  by  C.  E.  K.  Mees  at  the  Rochester 
meeting,  October- 12,  1936;  the  "Television  Demonstration"  by  the 
Radio  Corporation  of  America  at  the  New  York  meeting,  October  14, 
1937;  and  the  paper  on  "The  Transmission  of  Motion  Pictures  over 
a  Coaxial  Cable"  by  H.  E.  Ives  at  the  Washington  meeting,  April  25, 
1938. 

When  the  meetings  are  held  on  the  West  Coast,  however,  evening 


220  WORK  OF  PAPERS  COMMITTEE  [j.  S.  M.  P.  E. 

sessions  should  be  planned  because  many  of  the  technical  workers  in 
the  studios  find  it  difficult  to  get  away  from  production  during  the  day 
but  can  attend  an  evening  session.  Studio  visits  or  open  mornings 
should  be  arranged  to  permit  members  from  the  East  to  see  actual 
production  conditions. 

It  is  recommended  that  an  approximate  time  allotment  be  given 
each  paper  on  the  preliminary  program  so  the  author  will  know  the 
time  that  has  been  allowed  for  his  presentation  and  the  discussion  of 
it. 

The  preliminary  program  copy  should  be  sent  in  to  the  editor  suffi- 
ciently in  advance  of  the  meeting  to  permit  correction  of  proof  and 
mailing  of  the  program  at  least  three  weeks  before  ike  meeting, 

(6)  Mail  Preliminary  Program  to  Each  Author. — A  copy  of  the  pre- 
liminary program  accompanied  by  a  letter  should  be  mailed  to  every 
author.    This  letter  should  urge  the  author  to  condense  his  paper  and 
rehearse  its  presentation  to  keep  within  the  time  limits  specified. 
It  should  repeat  the  request  for  two  copies  of  the  manuscript  by  a 
specified  date.    A  sample  letter  is  appended  to  this  report.    A  colored 
paper  stock  helps  to  insure  that  the  recipient  reads  the  letter. 

(7)  Check  All  Manuscripts. — Every  manuscript  should  be  examined 
as  received  to  determine  whether  (1)  two  copies  have  been  sent  in; 
(2)  all  figures  are  included;    and  (3)  whether  drawings  and  graphs 
have  been  prepared  according  to  regulations,  etc.     If  some  of  these 
requirements  have  not  been  met,  the  author  should  be  infoimed  at 
once. 

Authors  who  have  failed  to  send  their  manuscripts  in  by  the  final 
date  specified  for  receipt  of  manuscripts  should  be  informed  by  letter 
that  the  Papers  Committee  desire  the  manuscript  at  the  earliest  con- 
venience of  the  author.  Papers  arriving  late  may,  at  the  discretion 
of  the  Committee,  be  scheduled  on  the  final  program  to  be  read  by 
title  or  substituted  for  other  papers  in  the  event  of  cancellations. 

It  should  also  be  pointed  out  to  authors  that  although  two  com- 
plete copies  of  their  manuscript  are  desired,  it  is  agreeable  to  supply 
preliminary  copies  requiring  further  slight  alterations  in  text  or  com- 
pletion of  illustrations  before  final  release.  Such  changes  should  be 
made  within  two  weeks  after  the  meeting. 

(8)  Preparation  of  Final  Program. — Copy  for  the  final  program 
should  be  prepared  about  one  week  before  the  meeting.     The  time 
for  delivery  of  each  paper  should  be  printed  along  the  left  margin. 
This  plan  has  several  advantages,  namely,  (1)  the  author  is  informed 


Feb.,  1939]  WORK  OF  PAPERS  COMMITTEE  221 

of  his  starting  time  and  total  time  for  delivery  and  discussion;  (2) 
those  members  and  guests  attending  the  meeting  know  approxi- 
mately which  paper  is  being  read  at  any  given  time ;  and  (3}  the  chair- 
man of  the  meeting  has  the  time  schedule  as  a  guide.  Whenever 
possible  the  arrangement  of  the  papers  on  the  final  program  should 
not  be  changed  from  that  of  the  preliminary  program.  Cancellations 
or  the  offer  of  a  paper  on  a  very  timely  subject  may  make  a  rearrange- 
ment necessary. 

In  general,  papers  whose  authors  will  not  be  present  should  usually  be 
placed  on  the  last  afternoon  or  at  the  end  of  the  session  on  other  days. 
Such  papers  should  be  marked  with  asterisks  with  the  explanation 
printed  that  papers  so  marked  will  be  restricted  to  ten  minutes  for 
presentation,  or  may  be  requested  to  be  read  by  title  if  the  time  is 
greatly  limited.  A  request  should  be  made  of  the  author  that  he 
assign  someone  to  give  a  digest  of  his  paper  and  inform  the  Papers 
Committee  several  days  before  the  meeting. 

Apparatus  papers  and  manufacturers'  announcements  of  new  prod- 
ucts should  generally  be  restricted  to  ten  minutes  for  presentation. 

The  proof  of  the  final  program  should  be  checked  by  the  Chairman 
on  Saturday  morning  before  the  Convention  opens  on  Monday.  A 
request  should  be  made  for  sufficient  copies  to  be  delivered  Saturday 
afternoon  or  evening  so  that  they  may  be  distributed  to  the  Board 
of  Governors  the  next  day  and  to  the  Publicity  Committee. 

(9)  Final  Check-  Up  at  Meeting. — It  is  important  that  the  Chairman 
of  the  Papers  Committee  get  in  touch  with  every  author  and  any 
other  individuals  who  have  been  assigned  to  read  papers  in  the  absence 
of  the  authors,  to  determine  whether  all  their  arrangements  are  com- 
plete and  that  they  are  ready  to  present  their  paper  at  and  for  the 
time  specified  on  the  final  program.    Good  showmanship  requires  that 
every  author  be  ready  when  called  upon  in  order  that  each  paper  will 
be  read  as  scheduled  on  the  program. 

The  question  of  undelivered  manuscripts,  missing  illustrations, 
corrections  on  manuscripts,  and  other  details  should  be  discussed 
with  the  author  or  his  designee  at  the  meeting  as  this  is  the  best  op- 
portunity for  the  Chairman  of  the  Committee  to  get  such  information. 

It  is  also  suggested  that  the  question  of  papers  for  the  next  meeting 
of  the  Society  be  kept  in  mind  and  that  suggestions  and  offers  for 
papers  be  recorded. 

(10)  Examination  of  Manuscripts  Preparatory  to  Submission  to  the 
Board  of  Editors. — A  final  check  should  be  made  of  every  manuscript, 


222  WORK  OF  PAPERS  COMMITTEE  [J.  S.  M.  P.  E. 

preferably  by  reading  the  manuscript  before  it  is  turned  over  to  the 
Chairman  of  the  Board  of  Editors  for  consideration  for  acceptance  for 
publication.  Notation  should  be  made  on  the  manuscript  of  any 
special  requests  made  by  the  author  relative  to  publication. 

General  Suggestions. — The  work  of  the  Papers  Committee  is  not  an 
easy  task  because  most  individuals  in  this  industry,  whether  in  re- 
search, manufacture,  production,  or  exhibition,  are  so  busy  with 
their  daily  problems  that  they  seldom  wish  to  take  the  time  to  write 
up  the  results  of  their  work.  The  most  interesting  papers,  however, 
are  often  those  dealing  with  current  developments  and,  if  the  Papers 
Committee  is  alert  to  its  responsibilities,  such  papers  can  only  be 
secured  by  approaching  those  individuals  who  are  working  in  the 
fields  in  question.  It  is  necessary,  therefore,  that  the  members  of  this 
Committee  establish  a  cordial  working  relationship  with  the  leaders 
in  various  centers  of  activity  in  the  industry.  When  these  leaders 
have  suitable  material  for  papers  for  our  Society,  they  are  more 
likely  to  offer  it  to  us  for  consideration.  These  leaders  should  be  cir- 
cularized at  intervals  as  to  the  possibility  of  papers  by  themselves  or- 
their  staffs. 

It  is  realized  that  it  is  not  possible  to  lay  down  any  set  of  rules  for 
the  most  satisfactory  organization  of  the  work  of  any  Committee 
but  it  is  hoped  that  the  suggestions  given  herein  may  prove  of  some 
value  to  those  individuals  who  accept  the  responsibility  of  chair- 
manship of  the  Papers  Committee  in  the  future. 

G.  E.  MATTHEWS 
Chairman,  Papers  Committee 


AUTHOR'S  FORM 
SMPE  FALL  CONVENTION 

Oct.  31-Nov.  2,  1938— Detroit,  Michigan. 

Please  fill  in  and  return  at  once  to Chairman,  Papers  Com- 
mittee, SMPE,  Kodak  Park,  Rochester,  N.  Y. 


1.  Title  of  Paper 

(Give  exact  wording) 

2.  Author  (s)  Name  (s) 

(Give  initials) 

3.  Company  Affiliation  and  Address 


Feb.,  1939]  WORK  OF  PAPERS  COMMITTEE  223 

4.  Abstract.    A  complete  abstract  (about  200  words)  is  required  by  Sept.  15th 
for  publication  in  the  October  number  of  the  Journal. 

5.  Manuscript.    It  is  requested  that  the  complete  manuscript  be  sent  in  to  the 
chairman  of  the  Papers  Committee  not  later  than  Sept.  15th  if  preferred  listing 
on  the  program  is  desired.    Two  copies  of  each  manuscript  must  be  received  by  October 
1st  or  the  paper  may  be  listed  to  be  read  by  title  on  the  final  program. 

6.  Do  you  expect  to  present  the  paper  in  person?     Yes No 

If  not,  who  will  present  the  paper? 

(Time  will  be  restricted  if  author  (s)  not  present) 

7.  Time  required  for  presentation Minutes 

(Usually  15-20  Min.) 

8.  Do  you  expect  to  show  lantern  slides?    Yes No 

(Facilities  will  be  provided) 

9.  Do  you  plan  to  show  films  with  your  papers?     Yes No 

Are  they  sound or  silent ?     Length? 35  mm.? 

(Ft.)         16mm.? 

10.  Special  Requirements.    State  in  detail  any  special  requirements  as  to  dem- 
onstrations, electrical  power  supply,  projection,  etc.     Facilities  will  be  provided 
for  the  projection  of  lantern  slides,  16-mm.  and  35-mm.  motion  picture  films. 


Rulings  on  Publicity  on  Papers  and  Acceptance  of  Papers  for  Publication. 

(1)  Acceptance  of  papers  by  the  Papers  Committee  does  not  imply  agreement  to 
publish.    The  Board  of  Editors  reserve  the  right  to  decline  to  publish  even  though 
the  paper  may  have  been  presented  at  the  convention,  unless  the  manuscript  is 
received  and  accepted  one  month  before  the  convention. 

(2)  Publicity  incident  to  the  presentation  of  papers  at  conventions  is  the  re- 
sponsibility solely  of  the  Papers  and  Publicity  Committees  of  the  Society  and 
should  not  be  undertaken  by  the  authors  or  their  representatives. 

(3)  For  further  details  on  rules  for  papers,  see  the  reprint  Regulations  of  the 
SMPE  Related  to  the  Preparation  of  Papers  for  Presentation  and  Publication. 


SPECIAL  BULLETIN  OF  THE  PAPERS  COMMITTEE 

OF    THE 
SOCIETY  OF    MOTION  PICTURE  ENGINEERS 


Place. 
Date. 


Name  and  Address  of  Author 

Dear  Mr 

A  preliminary  program  for  the  (Place)  meeting  is  enclosed  for  your  informa- 
tion. Please  note  the  date  and  time  specified  for  your  paper.  It  is  requested  that 
you  condense  your  paper  for  presentation  at  the  meeting  and  time  it  carefully  to 


224  WORK  OF  PAPERS  COMMITTEE 

fit  in  with  the  specified  time  requirements.  This  can  be  done  only  by  actual  re- 
hearsal. Papers  for  the  Apparatus  Symposium  will  be  restricted  to  10  minutes 
for  presentation. 

Papers  must  be  given  on  the  dates  specified,  unless  a  change  is  approved  by 
motion  of  the  convention  delegates  at  the  proposal  of  the  chairman. 

Papers  designated  with  an  asterisk  (*)  will,  in  the  absence  of  the  author  (s) 
generally  be  restricted  to  10  minutes  for  presentation  or  may  be  requested  to  be 
read  by  title,  if  the  time  is  greatly  limited. 

The  full  manuscript  should  be  submitted  for  publication  but  is  subject  to 
final  approval  by  the  Board  of  Editors.  Please  check  over  your  manuscript 
carefully  and  see  that  it  conforms  with  the  regulations  governing  papers. 

Two  copies  of  your  manuscript  must  be  in  my  hands  by  (Date)  in  order  that 
he  paper  be  listed  on  the  final  program. 

Yours  cordially, 

(Signature) 
Chairman,  Papers  Committee 


CURRENT  LITERATURE  OF  INTEREST  TO  THE  MOTION  PICTURE 

ENGINEER 


The  editors  present  for  convenient  reference  a  list  of  articles  dealing  with  subjects 
cognate  to  motion  picture  engineering  published  in  a  number  of  selected  journals. 
Photostatic  copies  may  be  obtained  from  the  Library  of  Congress,  Washington,  D.  C., 
or  from  the  New  York  Public  Library,  N.  Y.  Micro  copies  of  articles,  in  maga- 
zines that  are  available  may  be  obtained  from  the  Bibliofilm  Service,  Department  of 
Agriculture,  Washington,  D.  C. 


Educational  Screen 

17  (Nov.,  1938),  No.  9 
Motion  Pictures— Not  for  Theaters — III  (pp.  291-294). 

Electronics 

11  (Nov.,  1938),  No.  11 
Television  Synchronization  (pp.  18-20) 

An  Automatic  Remote  Amplifier  (p.  21). 
Selecting  Loud  Speakers  for  Special  Operating  Condi- 
tions (pp.  22-24). 
Light  and  Sound  (p.  25). 

A  Laboratory  Television  Receiver — V  (pp.  26-29). 
Advanced  Disc  Recording  (pp.  34-36,  82). 
International  Photographer 

10  (Nov.,  1938),  No.  10 
New  Stereoscopic  Method  (pp.  10-11). 
Projection  Symposium  (pp.  24-27). 
International  Projectionist 

13  (Nov.,  1938),  No.  11 

Advance  Preparations  Minimize  Sound  System  Emer- 
gencies (pp.  7-10). 
Mechanics  of  Motion  Picture  Projection  (pp.   10-11, 

21-26). 

Theater  Structure,  Screen  Light  and  Revised  Projection 
Room  Plans  (pp.  12-15). 

Kinematograph  Weekly 

261  (Nov.  3,  1938),  No.  1646 
Sound-Tracks  on  Ozaphane  Film  Stock  (p.  29). 
Motion  Picture  Herald,  Better  Theatres 

133  (Nov.  12,  1938),  No.  7 

Lighting  the  Theater  Interior  with  the  New  Fluorescent 
Lamps  (pp.  17-19,  25-26). 


A.  D.  KROWS 


E.  W.  ENGSTROM  and 
R.  S.  HOLMES 
G.  H.  BREWER 

L.  B.  HALLMAN,  JR. 

D.  G.  FINK 
C.  J.  LE  BEL 


L.  H.  SHIRPSER 
W.  S.  THOMPSON 


A.  NADELL 
J.  FRANK,  JR. 

Report  of  the  SMPE 
Projection  Practice 
Committee 


F.  M.  FALGE 


225 


1939  SPRING  CONVENTION 
SOCIETY  OF  MOTION  PICTURE  ENGINEERS 


HOLLYWOOD  ROOSEVELT  HOTEL 
HOLLYWOOD,   CALIFORNIA 
APRIL  17th-21st,  INCLUSIVE 


Officers  and  Committees  in  Charge 

E.  A.  WILLIFORD,  President 

N.  LEVINSON,  Executive  Vice-President 

W.  C.  KUNZMANN.  Convention  Vice-F 'resident 

J.I.  CRABTREE,  Editorial  V ice-President 

L.  RYDER,  Chairman,  Pacific  Coast  Section 

H.  G.  TASKER,  Chairman,  Local  Arrangements  Committee 

J.  HABER,  Chairman,  Publicity  Committee 

Pacific  Coast  Papers  Committee 

L.  A.  AICHOLTZ,  Chairman 

O.  O.  CECCARINI  W.  A.  MUELLER 

G.  CHAMBERS  H.  G.  TASKER 

L.  D.  GRIGNON  W.  H.  ROBINSON,  JR. 

C.  N.  BATSEL  C.  R.  SAWYER 

Reception  and  Local  Arrangements 

H.  G.  TASKER,  Chairman 

N.  LEVINSON  G.  F.  RACKETT  E.  HUSE 

K.  F.  MORGAN  H.  W.  MOYSE  L.  L.  RYDER 

P.  MOLE  W.  MILLER  J.  O.  AALBERG 

A.  M.  GUNDELFINGER       J.  A.  BALL  R.  H.  McCULLOUGH 

H.  W.  REMERSCHIED       W.  A.  MUELLER  J.  M.  NICKOLAUS 

G.  S.  MITCHELL  C.  L.  LOOTENS  E.  H.  HANSEN 

Registration  and  Information 

W.  C.  KUNZMANN,  Chairman 

S.  HARRIS  C.  W.  HANDLEY 

E.  R.  GEIB  W.  R.  GREENE 

Hotel  and  Transportation 

G.  A.  CHAMBERS,  Chairman 

W.  C.  HARCUS  C.  DUNNING  W.  E.  THEISEN 

G.  HOUGH  E.  C.  RICHARDSON  D.  P.  LOYE 

B.  KREUZER  K.  STRUSS  O.  O.  CECCARINI 
H.  W.  REMERSCHIED       J.  C.  BROWN  C.  J.  SPAIN 

226 


1939  SPRING  CONVENTION  227 

Convention  Projection 

H.  GRIFFIN,  Chairman 

].  O.  AALBERG  H.  A.  STARKE  J.  DURST 

C.  W.  HANDLEY  L.  E.  CLARK  R.  H.  McCuLLOUGH 

W.  F.  RUDOLPH  M.  S.  LESHING  H.  C.  SILENT 

J.  M.  NICKOLAUS  A.  F.  EDOUART  H.  I.  REISKIND 

J.  K.  HILLIARD  I.  SERRURIER  W.  W.  LINDSAY,  JR. 

Officers  and  Members  of  Los  Angeles  Projectionists  Local  No.  150 

Banquet  and  Dance 

N.  LEVINSON,  Chairman 

H.  T.  KALMUS  G.  S.  MITCHELL  G.  F.  RACKETT 

E.  HUSE  P.  MOLE  J.  O.  AALBERG 

L.  L.  RYDER  H.  G.  TASKER  K.  F.  MORGAN 

C.  DUNNING  W.  MILLER  H.  W.  MOYSE 

G.  A.  MITCHELL  R.  H.  MCCULLOUGH  J.  L.  COURCIER 

Ladies9  Reception  Committee 

MRS.  N.  LEVINSON,  Hostess 

assisted  by 

MRS.  E.  C.  RICHARDSON  MRS.  P.  MOLE  MRS.  E.  HUSE 

MRS.  G.  F.  RACKETT       MRS.  C.  W.  HANDLEY  MRS.  L.  L.  RYDER 

MRS.  H.  W.  MOYSE         MRS.  K.  F.  MORGAN  MRS.  J.  O.  AALBERG 

MRS.  H.  G.  TASKER        MRS.  W.  MILLER  MRS.  R.  H.  MCCULLOUGH 

MRS.  L.  E.  CLARK  MRS.  A.  M.  GUNDELFINGER  MRS.  C.  DUNNING 

Publicity 

J.  HABER,  Chairman 

L.  A.  AICHOLTZ  W.  R.  GREENE  S.  HARRIS 

W.  A.  MUELLER  G.  CHAMBERS  A.  M.  GUNDELFINGER 

New  Equipment  Exhibit 

J.  G.  FRAYNE,  Chairman 

P.  MOLE  C.  R.  DAILY  H.  W.  REMERSCHIED 

J.  DURST  S.  HARRIS  C.  N.  BATSEL 

O.  F.  NEV 

Headquarters 

Headquarters  of  the  Convention  will  be  the  Hollywood  Roosevelt  Hotel,  where 
excellent  accommodations  are  assured.  A  reception  suite  will  be  provided  for  the 
Ladies'  Committee,  and  an  excellent  program  of  entertainment  will  be  arranged 
for  the  ladies  who  attend  the  Convention. 

Special  hotel  rates,  guaranteed  to  SMPE  delegates,  European  plan,  will  be  as 
follows : 

One  person,  room  and  bath  $  3 . 50 

Two  persons,  double  bed  and  bath  5 . 00 

Two  persons,  twin  beds  and  bath  6 . 00 

Parlor  suite  and  bath,  1  person  8 . 00 

Parlor  suite  and  bath,  2  persons  12.00 


228  1939  SPRING  CONVENTION  fj.  s.  M.  P.  E. 

Indoor  and  outdoor  garage  facilities  adjacent  to  the  Hotel  will  be  available 
to  those  who  motor  to  the  Convention. 

Members  and  guests  of  the  Society  will  be  expected  to  register  immediately 
upon  arriving  at  the  Hotel.  Convention  badges  and  identification  cards  will 
be  supplied  which  will  be  required  for  admittance  to  the  various  sessions,  the 
studios,  and  several  Hollywood  motion  picture  theaters. 

Railroad  Fares 

The  following  table  lists  the  railroad  fares  and  Pullman  charges : 

Railroad 

Fare  Pullman 

City  (round  trip)  (one  way) 

Washington  $132.20  $22.35 

Chicago  90.30  16.55 

Boston  147.50  23.65 

Detroit  106.75  19.20 

New  York  139.75  22.85 

Rochester  124.05  20.50 

Cleveland  110.00  19.20 

Philadelphia  135.50  22.35 

Pittsburgh  117.40  19.70 

The  railroad  fares  given  above  are  for  round  trips,  sixty-day  limits.  Arrange- 
ments may  be  made  with  the  railroads  to  take  different  routes  going  and  coming, 
if  so  desired,  but  once  the  choice  is  made  it  must  be  adhered  to,  as  changes  in  the 
itinerary  may  be  effected  only  with  considerable  difficulty  and  formality.  Dele- 
gates should  consult  their  local  passenger  agents  as  to  schedules,  rates,  and  stop- 
over privileges. 

Technical  Sessions 

The  Hollywood  meeting  always  offers  our  membership  an  opportunity  to  be- 
come better  acquainted  with  the  studio  technicians  and  production  problems,  and 
arrangements  will  be  made  to  visit  several  of  the  studios.  The  Local  Papers 
Committee  under  the  chairmanship  of  Mr.  L.  A.  Aicholtz  is  collaborating  closely 
with  the  General  Papers  Committee  in  arranging  the  details  of  the  program. 
Complete  details  of  the  program  will  be  published  in  a  later  issue  of  the  JOURNAL. 

Semi- Annual  Banquet  and  Dance 

The  Semi- Annual  Banquet  of  the  Society  will  be  held  at  the  Hotel  on  Thursday, 
April  20th.  Addresses  will  be  delivered  by  prominent  members  of  the  industry, 
followed  by  dancing  and  entertainment.  Tables  reserved  for  8,  10,  or  12  persons; 
tickets  obtainable  at  the  registration  desk. 

New  Equipment  Exhibit 

An  exhibit  of  newly  developed  motion  picture  equipment  will  be  held  in  the 
Bombay  and  Singapore  Rooms  of  the  Hotel,  on  the  mezzanine.  Those  who  wish 
to  enter  their  equipment  in  this  exhibit  should  communicate  as  early  as  possible 
with  the  general  office  of  the  Society  at  the  Hotel  Pennsylvania,  New  York,  N.  Y. 


Feb.,  1939]  1939  SPRING  CONVENTION  229 

Motion  Pictures 

At  the  time  of  registering,  passes  will  be  issued  to  the  delegates  to  the  Conven- 
tion, admitting  them  to  the  following  motion  picture  theaters  in  Hollywood,  by 
courtesy  of  the  companies  named:  Grauman's  Chinese  and  Egyptian  Theaters 
(Fox  West  Coast  Theaters  Corp.),  Warner's  Hollywood  Theater  (Warner  Brothers 
Theaters,  Inc.),  Pantages  Hollywood  Theater  (Rodney  Pantages,  Inc.).  These 
passes  will  be  valid  for  the  duration  of  the  Convention. 

Inspection  Tours  and  Diversions 

Arrangements  are  under  way  to  visit  one  or  more  of  the  prominent  Hollywood 
studios,  and  passes  will  be  available  to  registered  members  to  several  Hollywood 
motion  picture  theaters.  Arrangements  may  be  made  for  golfing  and  for  special 
trips  to  points  of  interest  in  and  about  Hollywood. 

Ladies'  Program 

An  especially  attractive  program  for  the  ladies  attending  the  Convention  is 
being  arranged  by  Mrs.  N.  Levinson,  hostess,  and  the  Ladies'  Committee.  A 
suite  will  be  provided  in  the  Hotel,  where  the  ladies  will  register  and  meet  for 
the  various  events  upon  their  program.  Further  details  will  be  published  in  a 
succeeding  issue  of  the  JOURNAL. 

Points  of  Interest 

En  route:  Boulder  Dam,  Las  Vegas,  Nevada;  and  the  various  National  Parks. 

Hollywood  and  vicinity:  Beautiful  Catalina  Island;  Zeiss  Planetarium;  Mt. 
Wilson  Observatory;  Lookout  Point,  on  Lookout  Mountain;  Huntington  Li- 
brary and  Art  Gallery  (by  appointment  only);  Palm  Springs,  Calif.;  Beaches  at 
Ocean  Park  and  Venice,  Calif.;  famous  old  Spanish  missions;  Los  Angeles  Mu- 
seum (housing  the  SMPE  motion  picture  exhibit);  Mexican  village  and  street, 
Los  Angeles. 

In  addition,  numerous  interesting  side  trips  may  be  made  to  various  points 
throughout  the  West,  both  by  railroad  and  bus.  Among  the  bus  trips  available 
are  those  to  Santa  Barbara,  Death  Valley,  Agua  Caliente,  Laguna,  Pasadena, 
and  Palm  Springs,  and  special  tours  may  be  made  throughout  the  Hollywood 
area,  visiting  the  motion  picture  and  radio  studios. 

On  February  18,  1939,  the  Golden  Gate  International  Exposition  will  open 
at  San  Francisco,  an  overnight  trip  from  Hollywood.  The  Exposition  will  last 
throughout  the  summer  so  that  opportunity  will  be  afforded  the  eastern  members 
to  take  in  this  attraction  on  their  convention  trip. 


SOCIETY  ANNOUNCEMENTS 


ATLANTIC  COAST  SECTION 

As  a  result  of  recent  elections,  announced  at  the  meeting  of  the  Section  on 
January  llth,  the  following  members  constitute  the  Board  of  Managers  for  1939: 

*D.  E.  HYNDMAN,  Chairman 

G.  FRIEDL,  JR.,  Past-Chairman  **H.  GRIFFIN,  Manager 

*P.  J.  LARSEN,  Sec.-Treas.  *R.  O.  STROCK,  Manager 

*Term  expires  December  31,  1939 
**Term  expires  December  31,  1940 

The  meeting  of  January  llth  was  held  at  the  Hotel  Pennsylvania,  New  York, 
N.  Y.,  and  was  devoted  to  a  celebration  of  the  centenary  of  the  announcement  to 
the  French  Academy  of  Sciences  in  January,  1839,  by  Arago,  of  the  contributions 
of  Daguerre  to  the  art  of  photography.  Two  papers  were  presented  as  follows: 

"The  Early  History  of  Photography,"  by  Edward  Epstean. 
"Daguerre's  Contribution  to  Photography,"  by  Beaumont  Newhall. 

The  joint  presentation  told  the  story  of  how  the  world  was  awakened  in  1839  to 
the  idea  of  photography  and  how,  as  the  result  of  the  divulgation  of  Daguerre's 
method,  scientists  were  stimulated  to  perfect  their  independent  processes,  notably 
John  Fox  Talbot  in  England,  whose  work  was  concerned  with  a  positive-negative 
process  admitting  of  duplication  by  printing,  whereas  Daguerre's  was  a  "one-shot" 
method. 

The  details  of  the  birth  of  photography  and  the  early  technics  were  also  de- 
scribed, and  Mr.  Newhall's  lecture  was  illustrated  by  several  lantern-slide  copies 
of  original  Daguerreotypes.  The  meeting  was  well  attended  and  considerable 
interest  was  shown  in  the  presentations. 

HOLLYWOOD   CONVENTION 

As  outlined  in  the  preceding  section  of  this  issue,  and  also  as  announced  on  the 
inside  front  cover,  the  next  convention  of  the  Society  will  be  held  on  April  17th- 
21st,  inclusive,  at  Hollywood,  Calif.,  with  headquarters  at  the  Hollywood  Roose- 
velt Hotel. 


230 


SOCIETY  OF  MOTION  PICTURE  ENGINEERS 

REPORT  OF  THE  TREASURER  FOR  1938 


Balance,  Dec.  31,  1937 
Receipts  during  1938 
Membership  dues 
Sustaining  Membership 
Publication   (Journal  sales,  reprints, 

subscriptions,  advertising,  etc.} 
Other    income    (membership    certifi- 
cates,  Journal  binders,   test-films, 
interest,  etc.} 

Total 

Disbursements  during  1938 

Publication  (Journal,  reprints,  bind- 
ers, etc.} 

Office  expenses,  rent,  and  salaries 

Officers'  expenses 

Local  Sections 

Other  expenses  (dues  and  fees,  test- 
films,  misc.) 

Total 


$21,353.84 


$12,217.94 
3,400.00 

5,008.84 


9,860.34 


$30,487.12 


$10,810.45 

10,670.13 

1,014.62 

655.30 

4,466.60 


$27,617.10 


2,870.02 


Balance,  December  31,  1938 


$24,223.  J 
L.  W.  DAVEE,  Treasurer 


231 


BACK  NUMBERS  OF  THE  TRANSACTIONS  AND  JOURNALS 

Prior  to  January,  1930,  the  Transactions  of  the  Society  were  published  quar- 
terly. A  limited  number  of  these  Transactions  are  still  available  and  will  be 
sold  at  the  prices  listed  below.  Those  who  wish  to  avail  themselves  of  the  op- 
portunity of  acquiring  these  back  numbers  should  do  so  quickly,  as  the  supply 
will  soon  be  exhausted,  especially  of  the  earlier  numbers.  It  will  be  impossible 
to  secure  them  later  on  as  they  will  not  be  reprinted. 


1920 
1921 
1922 

1924 


No. 

10 

12 

15 

f!9 

\20 


Price 

$1.00 
1.00 
1.00 
1.25 
1.25 


1925 


1926 


No. 
21 

22 
23 
24 
25 
26 
27 
28 


Price 
$1.25 
1.25 
1.25 
1.25 
1.25 
1.25 
1.25 
1.25 


1927 


1928 


1929 


No. 
29 
32 
33 
34 
35 
36 
37 
38 


Price 

$1.25 
1.25 
2.50 
2.50 
2.50 
2.50 
3.00 
3.00 


Beginning  with  the  January,  1930,  issue,  the  JOURNAL  of  the  Society  has  been 
issued  monthly,  in  two  volumes  per  year,  of  six  issues  each.  Back  numbers  of 
all  issues  are  available  at  the  price  of  $1.00  each,  a  complete  yearly  issue  totalling 
$12.00.  Single  copies  of  the  current  issue  may  be  obtained  for  $1.00  each. 
Orders  for  back  numbers  of  Transactions  and  JOURNALS  should  be  placed  through 
the  General  Office  of  the  Society  and  should  be  accompanied  by  check  or 
money-order. 

SOCIETY  SUPPLIES 


The  following  are  available  from  the  General  Office  of  the  Society,  at  the  prices 
noted.  Orders  should  be  accompanied  by  remittances. 

Aims  and  Accomplishments. — An  index  of  the  Transactions  from  October, 
1916,  to  December,  1929,  containing  summaries  of  all  articles,  and  author  and 
classified  indexes.  One  dollar  each. 

Journal  Index. — An  index  of  the  JOURNAL  from  January,  1930,  to  December, 
1935,  containing  author  and  classified  indexes.  One  dollar  each. 

SMPE  Standards.— The  revised  edition  of  the  SMPE  Standards  and  Recom- 
mended Practice  was  published  in  the  March,  1938,  issue  of  the  JOURNAL,  copies 
of  which  may  be  obtained  for  one  dollar  each. 

Membership  Certificates. — Engrossed,  for  framing,  containing  member's  name, 
grade  of  membership,  and  date  of  admission.  One  dollar  each. 

Lapel  Buttons. — The  insignia  of  the  Society,  gold  filled,  with  safety  screw  back. 
One  dollar  each. 

Journal  Binders. — Black  fabrikoid  binders,  lettered  in  gold,  holding  a  year's 
issue  of  the  JOURNAL.  Two  dollars  each.  Member's  name  and  the  volume 
number  lettered  in  gold  upon  the  backbone  at  an  additional  charge  of  fifty  cents 
each. 

Test- Films. — See  advertisement  in  this  issue  of  the  JOURNAL. 


JOURNAL 

OF  THE  SOCIETY  OF 

MOTION  PICTURE  ENGINEERS 

Volume  XXXII  March,  1939 


CONTENTS 

Page 

Latest  Developments  in  Variable- Area  Processing 

A.  C.  BLANEY  AND  G.  M.  BEST    237 

Improving  the  Fidelity  of  Disk  Records  for  Direct  Playback .  . 

H.  J.  HASBROUCK    246 

The  Centenary  of  Photography  and  the  Motion  Picture 

EDWARD  EPSTEAN    253 

The  Surface  of  the  Nearest  Star R.  R.  McMATH     264 

The  Electrical  Production  of  Musical  Tones S.  T.  FISHER     280 

A  Color-Temperature  Meter 

E.  M.  LOWRY  AND  K.  S.  WEAVER     298 

Chemical  Analysis  of  an  MQ  Developer 

R.  M.  EVANS  AND  W.  T.  HANSON,  JR.     307 

An  Opacimeter  Used  in  Chemical  Analysis 

R.  M.  EVANS  AND  G.  P.  SILBERSTEIN    321 

A  New  Projector  Mechanism H.  GRIFFIN  325 

1939  Spring  Convention  at  Hollywood,  Calif 336 

Society  Announcements 341 

Constitution  and  By-Laws  of  the  Society 344 


JOURNAL 

OF  THE  SOCIETY  OF 

MOTION  PICTURE  ENGINEERS 


SYLVAN  HARRIS,  EDITOR 

Board  of  Editors 
J.  I.  CRABTREE,  Chairman 

A.  N.  GOLDSMITH  A.  C.  HARDY  H.  G.  KNOX 

J.  G.  FRAYNE  L.  A.  JONES  G.  E.  MATTHEWS 

E.  W.  KELLOGG 


Subscription  to  non-members,  $8.00  per  annum;  to  members,  $5.00  per  annum, 
included  in  their  annual  membership  dues;  single  copies,  $1.00.  A  discount 
on  subscription  or  single  copies  of  15  per  cent  is  allowed  to  accredited  agencies. 
Order  from  the  Society  of  Motion  Picture  Engineers,  Inc.,  20th  and  Northampton 
Sts.,  Easton,  Pa.,  or  Hotel  Pennsylvania,  New  York,  N.  Y. 
Published  monthly  at  Easton,  Pa.,  by  the  Society  of  Motion  Picture  Engineers. 

Publication  Office,  20th  &  Northampton  Sts.,  Easton,  Pa. 
General  and  Editorial  Office,  Hotel  Pennsylvania,  New  York,  N.  Y. 

West-Coast  Office,  Suite  226,  Equitable  Bldg.,  Hollywood,  Calif. 
Entered  as  second  class  matter  January  15,  1930,  at  the  Post  Office  at  Easton, 
Pa.,  under  the  Act  of  March  3,  1879.     Copyrighted,  1939,  by  the  Society  of 
Motion  Picture  Engineers,  Inc. 

Papers  appearing  in  this  Journal  may  be  reprinted,  abstracted,  or  abridged 
provided  credit  is  given  to  the  Journal  of  the  Society  of  Motion  Picture  Engineers 
and  to  the  author,  or  authors,  of  the  papers  in  question.  Exact  reference  as  to 
the  volume,  number,  and  page  of  the  Journal  must  be  given.  The  Society  is  not 
responsible  for  statements  made  by  authors. 


OFFICERS  OF  THE  SOCIETY 

**  President:     E.  A.  WILLIFORD,  30  East  42nd  St.,  New  York,  N.  Y. 
**  Past-President:     S.  K.  WOLF,  RKO  Building,  New  York,  N.  Y. 
**  Executive  Vice-P resident:     N.  Levinson,  Burbank,  Calif. 

*  Engineering  Vice-P  resident:     L.  A.  JONES,  Kodak  Park,  Rochester,  N.  Y. 
**  Editorial  Vice-President:    J.  I.  CRABTREE,  Kodak  Park,  Rochester,  N.  Y. 

*  Financial  Vice-President:    A.  S.  DICKINSON,  28  W.  44th  St.,  New  York,  N.  Y. 
**  Convention  Vice-President:    W.  C.  Kunzmann,  Box  6087,  Cleveland,  Ohio. 

*  Secretary:    J.  FRANK,  JR.,  90  Gold  St.,  New  York,  N.  Y. 

*  Treasurer:     L.  W.  DAVEE,  153  Westervelt  Ave.,  Tenafly,  N.  J. 

GOVERNORS 
**  M.  C.  BATSEL,  Front  and  Market  Sts.,  Camden,  N.  J. 

*  R.  E.  FARNHAM,  Nela  Park,  Cleveland,  Ohio. 

*  H.  GRIFFIN,  90  Gold  St.,  New  York,  N.  Y. 

*  D.  E.  HYNDMAN,  350  Madison  Ave.,  New  York,  N.  Y. 

*  L.  L.  RYDER,  5451  Marathon  St.,  Hollywood,  Calif. 

*  A.  C.  HARDY,  Massachusetts  Institute  of  Technology,  Cambridge,  Mass. 

*  S.  A.  LUKES,  6427  Sheridan  Rd.,  Chicago,  111. 

**  H.  G.  TASKER,  14065  Valley  Vista  Blvd.,  Van  Nuys  Calif. 

*  Term  expires  December  31,  1939. 
**  Term  expires  December  31,  1940. 


LATEST  DEVELOPMENTS  IN  VARIABLE-AREA 
PROCESSING* 


A.  C.  BLANEY**  AND  G.  M.  BESTf 


Summary. — The  purpose  of  this  paper  is  to  present  a  series  of  curves  showing  the 
photographic  control  oj  variable-area  sound-tracks  in  commercial  production  at 
Warner  Bros.  Studio,  and  to  show  the  wide  tolerances  in  film  processing  permissible 
with  Class  A  push-pull  recording,  a  factor  that  is  of  especial  interest  in  daily  produc- 
tion. 

The  results  of  a  study  of  the  technic  involved  in  fine-grain  photographic  duplicating 
of  variable-area  sound-track,  for  foreign  release  are  also  discussed. 

The  photographic  control  of  variable-area  sound-tracks  in  com- 
mercial laboratories  has  been  satisfactorily  established  by  the  use  of 
modulated  high-frequency  test  recordings.  The  nature  of  these  tests 
has  been  previously  described  by  Baker  and  Robinson  r1 

The  quality  of  variable-width  sound  records  depends  to  a  great  extent  upon 
image  definition.  The  requirements  for  a  perfect  sound-track  are  complete 
transparency  in  the  clear  portion,  complete  opacity  in  the  dark  portions,  an 
extremely  sharp  boundary  between  the  clear  and  dark  portions,  and  exact  dupli- 
cation of  the  wave  traced  upon  the  track  by  the  galvanometer. 

Distortion  is  introduced  by  any  change  in  average  transmission  in  recording 
high-frequency  waves.  At  high  densities  the  average  transmission  is  reduced, 
and  at  very  low  densities  is  increased  by  the  presence  of  the  high-frequency  waves. 
The  average  transmission  is  compared  to  the  transmission  through  the  film  for  a 
50-per  cent  exposed  track  without  signal. 

It  is  possible  to  find  a  density  at  which  there  is  little,  if  any,  change  in  average 
transmission,  and  this  density  corresponds  to  most  nearly  perfect  image  defini- 
tion and  least  distortion  .... 

A  modulated  high-frequency  recording  affords  an  extremely  accurate  means  of 
determining  correct  negative  and  print  densities  for  given  conditions  of  labora- 
tory processing.  An  oscillator,  designed  for  several  carrier  frequencies,  is  pro- 
vided with  a  400-cycle  modulator  for  recording.  The  modulated  carrier  is 
recorded  for  several  values  or  lamp  current,  and  processed  to  several  negative 
densities.  Prints  are  then  proceeded  to  various  values  of  density,  and  the  400- 

*Presented  at  the  1938  Fall  Meeting  at  Detroit,  Mich.;   received  October 
10,  1938. 

**RCA  Manufacturing  Co.,  Hollywood,  Calif. 
fWarner  Bros   Pictures,  Inc.,  Burbank,  Calif. 

237 


238 


A.  C.  BLANEY  AND  G.  M.  BEST 


[J.  S.  M.  P.  E. 


^ 


J:     f 


:*   H 

I    -8 

Q-  hfi 

I  .fH 

to 


oij.v-\oaow  SSOSD  do 


V 

HtCA,TWt  DtHSHY 

FIG.  1.  Cancellation  curves  for  standard  bi- 
iteral  and  Class  A  push-pull  tracks,  plotted 
gainst  negative  density. 

1 

>v 

"a.  \^ 

% 

\ 

5§ 

^ 

^s 

IP 

"~" 

A 

^ 

^  

,  * 

x" 

/ 

/ 

/ 

/ 

7 

?         ?        ?        a        a        s         is        §        ? 

Mar.,  1939]  VARIABLE -AREA  PROCESSING  239 

cycle  output  measured  on  suitable  reproducing  equipment.  The  combination  of 
negative  and  print  densities  that  gives  least  400-cycle  output  indicates  the  con- 
dition for  best  image  definition  and  least  distortion  .... 

This  method  of  test  was  used  to  determine  the  data  presented  in 
this  paper. 

In  the  daily  production  of  sound-tracks,  certain  variations  in  track 
density  are  always  present.  These  variations  are  due  to  a  number 
of  causes,  such  as  emulsion  speed,  exposure,  development,  retrogres- 
sion, temperature,  etc.  While  these  variations  may  be  held  to  a 
minimum,  they  become  apparent  when  the  final  film  is  assembled  for 
re-recording.  Such  variations  require  accurate  measurement  of  track 
density,  the  use  of  a  card-index  system  for  indexing  the  densities  of 
the  several  hundred  scenes  of  each  production,  and  the  timing  of  the 
cut  negative  when  prints  for  re-recording  purposes  are  made.  This 
procedure  involves  clerical  expense,  extra  handling  of  the  negative, 
and  a  considerable  loss  of  time  in  preparing  the  re-recording  prints. 

Such  variations  are  practically  eliminated  by  the  use  of  Class  A 
push-pull  for  the  original  daily  production.  The  advantage  of  the 
Class  A  track  is  primarily  due  to  cancellation  of  even-harmonic  dis- 
tortions which  may  arise  from  daily  variations  in  the  process.  Thus 
the  daily  track,  recorded  push-pull,  provides  a  higher  average  and 
more  uniform  quality  than  the  standard  track.  Accurate  timing  of 
the  negative  is  unnecessary  when  making  the  print  for  re-recording, 
as  variations  in  negative  density  within  reasonable  limits  produce  so 
little  change  in  the  relative  level  of  cross-modulation  products  that 
they  can  be  disregarded. 

Fig.  1  shows  the  comparative  cancellation  curves  for  standard 
bilateral  and  Class  A  push-pull  tracks  plotted  against  negative  den- 
sity. These  data  are  established  by  frequency  tests  consisting  of  (1) 
1000  cycles  for  reference  level,  (2)  modulated  9000  cycles  for  distor- 
tion test.  It  has  been  found  by  practical  experience  that  a  cancella- 
tion of  30  db.  is  satisfactory  for  all  types  of  material;  therefore  den- 
sity tolerances  may  be  established  at  this  value  of  cancellation.  Thus 
from  Fig.  1  the  negative  density  tolerance  for  the  standard  track 
at  a  print  density  of  1.59  is  from  1.93  to  2.10,  while  for  the  push-pull 
track  the  density  latitude  is  almost  unlimited.  It  can  be  readily  seen 
from  this  that  variations  in  negative  density  present  in  the  daily  pro- 
duction become  relatively  unimportant  when  using  the  push-pull 
system,  and  that  only  very  erratic  conditions  would  necessitate  special 
timing  of  the  negative. 


240  A.  C.  BLANEY  AND  G.  M.  BEST  [j.  s.  M.  p.  E. 

Fig.  2  shows  the  same  comparison,  plotted  against  print  density. 
Here  the  optimum  negative  density  as  indicated  in  Fig.  1  is  made  into 
a  loop  and  printed  to  a  number  of  densities  within  the  range  of  the 
printer.  Again  the  latitude  in  print  density  for  the  standard  track 
is  somewhat  limited,  being  within  the  range  of  1.53  to  1.70,  whereas 
the  push-pull  is  almost  unlimited.  While  the  latitude  of  the  standard 
track  is  sufficiently  broad  to  cover  variations  in  the  re-recorded  nega- 
tive and  release  printing,  the  use  of  push-pull  tremendously  simplifies 
the  daily  production  routine. 

Due  to  theater  limitations,  it  is  necessary  to  make  all  re-recorded 
tracks  standard  for  release.  However,  since  the  re-recording  is  ac- 
complished in  a  comparatively  short  period  of  time,  is  done  in  large 
sections,  and  is  subject  to  better  control  than  the  daily  production, 
the  variations  are  much  less.  The  density  tolerances  on  the 
standard  track  are  sufficiently  wide  to  accommodate  all  normal 
variations  that  exist  during  the  re-recording  process. 

Photographic  Dupes. — Most  producing  companies  retain  the  original 
sound-track  and  picture  negatives  in  the  United  States,  preferring  to 
send  photographic  dupes  to  the  foreign  market  rather  than  risk  having 
the  original  negative  cut  by  censors  or  damaged  in  transit.  For  eco- 
nomic reasons,  most  photographic  dupes  for  foreign  release  are  made 
with  the  picture  and  sound-track  on  the  same  film.  Since  consider- 
able emphasis  has  been  placed  on  the  necessity,  of  having  a  gamma 
of  2.00  or  over  for  variable-area  tracks,  there  has  been  much  skep- 
ticism as  to  the  possibility  of  making  high-quality  composite  dupes 
by  following  the  picture  process,  which  requires  the  use  of  relatively 
low  gammas.  It  must  be  pointed  out  that  high  gamma  is  necessary 
only  when  it  must  be  used  as  a  means  to  increase  sharpness;  if  the 
emulsion  and  development  produce  the  necessary  contrast  and  sharp- 
ness, the  actual  gamma  is  unimportant. 

With  fine-grain  duplicating  positive  and  negative  emulsions  of 
great  resolving  power  now  available,  experiments  were  first  carried 
out  in  the  RCA  engineering  laboratory  at  Camden  to  determine  what 
degree  of  sound  quality  could  be  reproduced  in  a  composite  photo- 
graphic dupe.  Frequency  measurements  indicated  the  losses  to  be 
very  small,  and  duplicate  prints  of  speech  and  music  compared  very 
favorably  with  prints  from  the  original  negative.  Production  work 
at  the  Warner  Hollywood  laboratory  proved  conclusively  the  satis- 
factory operation  of  the  process.  All  data  for  this  paper  were  taken 
from  production  work  at  the  Warner  Bros,  laboratory. 


Mar.,  1939]  VARIABLE- ARE  A  PROCESSING  241 

Since  the  picture  specifications  control  the  development  character- 
istic of  each  step  of  the  duplicating  process,  there  is  left  only  the 
quality  and  intensity  of  the  printing  light  to  be  controlled  in  printing 
the  track. 

The  frequency  negative  for  the  duplicating  tests  is  made  on  the 
same  recorder  and  at  the  same  time  as  the  re-recorded  negative,  so 
that  it  will  exactly  represent  the  conditions  of  the  release  negative. 
This  negative  is  also  used  for  control  of  the  domestic  release  prints. 

Fig.  3  shows  the  cancellation  curve  for  the  Eastman  fine-grain  du- 
plicating positive  stock  type  1365  developed  to  a  gamma  of  1.26  in 
D-76  developer,  printed  from  the  release  negative  test  having  a  dens- 
ity of  2.05.  This  print  is  exposed  with  white  light  because  there  is 
very  little  sharpness  to  be  gained  by  using  filtered  light  on  this  emul- 
sion. It  is  seen  that  the  greatest  cancellation  occurs  at  a  positive 
density  of  1.45,  which,  if  the  print  was  to  be  used  for  theater  reproduc- 
tion, would  be  the  correct  print  density.  However,  it  is  not  desirable 
to  use  this  print  density  for  making  the  duplicate  negative.  Due  to 
picture  specifications,  the  dupe  negative  is  developed  to  a  low  gamma 
and  will  thus  have  a  comparatively  large  amount  of  image  spread. 
Therefore,  to  cancel  some  of  this,  it  is  desirable  to  use  a  master 
positive  having  image  spread  in  the  opposite  direction.  A  density  of 
1.9  on  the  type  1365  fine-grain  positive  has  considerable  image  spread 
and  is  about  the  highest  density  contrast  that  can  be  obtained  on  the 
track  at  this  gamma,  so  a  print  density  of  this  value  is  used  for  the 
master  positive. 

Fig.  4  is  the  cancellation  curve  of  the  dupe  negative  printed  from 
the  1.9-derisity  master.  The  stock  used  is  the  Eastman  fine-grain 
duplicating  negative  type  1203  developed  in  D-76  developer  to  a 
gamma  of  0.58.  This  stock  is  panchromatic  in  its  sensitivity,  and 
considerable  gain  in  sharpness  can  be  obtained  by  the  use  of  a  filter. 
A  Corning  No.  556  filter  5  mm.  thick  restricts  the  actinic  light  to 
wavelengths  shorter  than  5000  A  and  procures  most  of  the  possible 
advantage.  Also,  the  total  amount  of  light  required  for  printing  this 
stock  with  the  filter  is  no  more  than  is  required  for  the  type  1365  stock. 

The  curve  shows  that  the  maximum  cancellation  occurs  at  the 
same  density  as  the  maximum  density  contrast;  therefore  a  density 
of  1.33  is  indicated  for  making  the  final  prints.  However,  it  has  been 
found  from  experience  with  a  number  of  pictures  put  through  the 
duping  process,  that  it  is  advisable  to  work  within  a  density  range  on 
the  dark  side  of  the  indicated  maximum  cancellation  point,  so  that 


242 


A.  C.  BLANEY  AND  G.  M.  BEST 


[J.  S.  M.  P.  E. 


t.o  I.I  I.Z  I.S  1.4  1-5  >«  1-7 
DUPE  hJtqATlVE  DENSITY  ^B*>5E 

FIG.  4.  Cancellation  curve  of  dupe  negative 
printed  from  the  1.9-density  master:  fine-grain 
duplicating  negative  type  1203,  D-76  developer, 
gamma  0.58. 

NI  Cl^X'Bd 

HJ 

Ob 

r*3s*5 

^^>- 

\ 

°-^ 

*tf~***\ 

\ 

x^ 

--—  -^ 

r 

,         T7r7?rr^. 

a  a  NI  sa.:>oaoa<i  Noirvioaow  SSOSD  jo  naA'ai  •aAu.vra* 

\ 

I.I  1.2.  i.5  14  i.S  1.6  1.7  1.6  )•<*  z. 

M*»ftR.  PoeiTWt  OtKlSlTV 

FIG.  3.  Cancellation  curve  of  EK  fine-grain 
duplicating  positive  stock  type  1365  (gamma  1.26, 
D-76  developer),  release  negative  test  density  2.05. 

\ 

£ 
\ 

\ 

^X 

\x 

^-~^ 

^5*1 

> 

^- 

-—  > 

/ 

/ 

/ 

Dtnoin                               in              o              in              o              io 

i           77€y1jj*>*>^->«- 

a  a  NI  siooaottd  Noii.vif\aow  csoaa  ^o  TIAIT  -jAu>n« 

Mar.,  1939] 


VARIABLE-AREA  PROCESSING 


243 


in  commercial  practice  this  range  is  from  1.33  to  1.40.  Negatives  that 
are  below  the  1.33  optimum  density  quickly  become  excessively  sibi- 
lant, and  those  that  are  kept  within  the  range  from  1.33  to  1.40  suffer 
no  impairment  of  frequency  characteristic  and  no  sibilants  are  intro- 
duced in  excess  of  those  already  existent  in  the  recording. 

Fig.  5  shows  the  cancellation  curve  for  the  positive  type  1301 
prints  made  from  the  dupe  negative.    These  prints  are  made  accord- 


-5 


-30 


g 

s- 

i 

-45 


\ 


"f 


I.O          I.I  l.Z  1.3          1.4  IS 

DOPHCATt  PRINT  DENSITY 


FIG.  5.     Cancellation  curve  for  positive  type  prints 
made  from  the  dupe  negative. 

ing  to  the  standard  release  practice  restricting  the  exposing  light  to  a 
wavelength  shorter  than  4000  A  by  a  Corning  584  filter  and  developed 
to  a  gamma  of  2.28.  The  maximum  cancellation  value  as  indicated 
by  the  curve  at  a  track  density  of  1.35  is  used  for  the  final  prints  from 
the  duplicate  negative.  The  curve  also  indicates  a  possible  ±0.1 
density  variation  without  serious  effect. 

Frequency  response  tests  indicate  an  overall  level  loss  of  1  db.  due 
to  the  printed-in  fog  on  the  dupe  prints,  and  an  attenuation  at  7000 
cycles  of  1  db.,  as  compared  with  an  original  print. 


244  A.  C.  BLANEY  AND  G.  M.  BEST  [j.  s.  M.  P.  E. 

The  sound  quality  that  can  be  obtained  in  a  print  is  largely  depend- 
ent upon  the  performance  of  the  printer.  Slippage  must  either  be 
eliminated  or  reduced  to  a  minimum  and  held  constant.  And,  most 
important  of  all,  the  films  must  be  maintained  in  absolute  contact. 
The  benefit  of  these  factors  can  be  appreciated  only  by  experiencing 
the  results  produced  by  such  a  printer. 

It  is  advisable  to  carry  the  frequency  test  through  the  duping  proc- 
ess along  with  the  production  material.  This  test  becomes  useful  in 
two  ways :  first,  it  is  a  check  on  the  duping  process,  and  second,  the 
test  dupe  negative  is  sent  to  the  foreign  laboratory  so  that  the  proper 
printing  conditions  may  be  established  for  that  particular  production. 

A  number  of  pictures  have  been  satisfactorily  duped  by  this  process 
at  the  Warner  Hollywood  laboratory.  It  is  their  practice  to  make 
one  master  positive  from  which  two  dupe  negatives  are  made.  All 
these  films  are  exported  to  foreign  markets.  However,  in  order  to 
check  the  process  a  complete  composite  release  print  is  made  from 
each  dupe  negative.  As  evidence  of  the  quality  obtained,  these  prints 
compare  so  favorable  with  those  made  direct  from  the  original  nega- 
tive that  they  are  used  in  the  domestic  release. 

REFERENCE 

1  BAKER,  J.  O.,  AND  ROBINSON,  D.  H.:  "Modulated  High-Frequency  Record- 
ing as  a  Means  of  Determining  Conditions  for  Optimal  Processing,"  /.  Soc.  Mot. 
PicL  Eng.,  XXX  (Jan.,  1938),  p.  3. 

DISCUSSION 

MR.  ROBERTS:  Why  in  all  these  curves  did  the  authors  obtain  optima  on  both 
negative  and  positive  processes?  It  is  my  understanding  of  cross-modulation 
theory  that  a  certain  fill  occurs  in  the  negative  process,  and  then  a  certain  amount 
of  fill  of  the  opposite  kind  in  the  positive  process,  which  causes  a  dip  in  the  curve 
forming  an  optimum  position  which  we  try  to  print. 

In  this  process  the  authors  apparently  get  an  optimum  on  1365  positive  film. 
One  would  expect  that,  because  the  fill  of  the  negative  is  being  cancelled.  Now, 
they  pick  a  point  above  the  optimum,  where  the  cancellation  is  greater  than  is 
needed.  Therefore,  one  would  expect  that  the  negative  on  1203  film  would  give 
an  optimum;  but  the  authors  apparently  pick  the  optimum  point  on  the  1203 
negative  curve  and  make  their  prints  from  that. 

It  would  seem  to  me  that  one  would  be  finished  once  he  had  picked  an  optimum 
for  the  negative,  and  not  expect  an  optimum  on  the  final  1301  print,  but  just  get  a 
sort  of  rising  curve. 

MR.  WOLFE:  No  matter  what  the  image-spread  may  be  on  the  negative  with 
which  we  start,  there  is  an  optimum  print  from  that  negative.  It  may  not  always 
be  the  same.  We  do  normally  try  to  select  the  density  of  the  original  negative  at  a 


Mar.,  1939]  VARIABLE-AREA  PROCESSING  245 

point  that  will  bring  the  cancellation  at  a  desired  density  on  the  print,  but  if  we 
use  a  different  negative  density  there  would  still  be  an  optimum  print  density,  but 
not  at  the  same  point. 

MR.  ROBERTS:  In  using  a  negative  and  print  we  have  a  combination.  In 
general,  we  get  optima  on,  say,  print  processes,  but  why  is  it  you  get  optima  in  both 
negative  and  positive?  I  regard  the  negative  and  positive  as  a  unit,  and  we  get  a 
family  of  curves  for  every  negative  and  positive.  But  if  we  plotted  the  cross- 
modulation  products  of  the  original  negative,  as  it  is  taken  from  the  developing 
machine,  we  would  not  get  a  minimum.  Why,  then,  should  we  get  a  minimum  in 
a  duplicate  negative  process? 

MR.  WOLFE:  If  you  take  a  negative  and  vary  the  exposure  you  will  get  a 
minimum.  We  sometimes  make  use  of  that  fact  when  we  wish  to  play  the  nega- 
tive. For  example,  if  we  know  in  advance  that  a  particular  piece  of  film  is  being 
made  for  play-back  purposes  only,  it  will  not  be  printed,  and  it  is  exposed  differ- 
ently from  the  way  in  which  it  would  be  exposed  if  we  expected  to  print  it ;  so  the 
variable  is  exposure  on  the  negative.  Again  I  think  the  point  is  quite  definite, 
that  there  is  a  minimum  in  every  case,  and  where  that  minimum  occurs  depends 
entirely  upon  the  preceding  process. 

MR.  ROBERTS :  I  had  the  idea  that  in  one  process  we  got  only  one  sort  of  effect ; 
that  is,  in  the  negative  we  have  only  the  fill  of  the  valley;  and  in  the  print  we  fill 
the  valley  corresponding  to  the  peak  of  the  negative,  so  as  to  equalize  and  make 
the  sound-wave  symmetrical. 

MR.  WOLFE:  What  we  are  talking  about  here  is  image-spread,  and  it  must  be 
clear  that  in  every  photographic  process  there  is  an  exposure  and  a  development 
condition  that  results  in  a  minimum  amount  of  image-spread.  It  so  happens  that 
for  the  normal  practical  process  we  expose  and  develop  the  negative  in  a  manner 
that  does  not  give  the  negative  minimum  image-spread. 

That  is  done  in  order  that  the  image-spread  may  be  used  to  cancel  the  image- 
spread  of  the  print  that  results  when  the  print  is  at  a  density  value  desirable  from 
the  standpoint  of  the  level  of  sound  reproduced. 


IMPROVING  THE  FIDELITY  OF  DISK  RECORDS  FOR 
DIRECT  PLAYBACK* 


H.  J.  HASBROUCK** 

Summary. — Recent  advances  in  design,  and  in  materals  of  which  the  recording 
disks  are  composed,  have  resulted  in  improved  fidelity.  Both  the  volume  range  ob- 
tainable, and  the  frequency  range,  have  been  extended,  satisfying  present-day  require- 
ments of  motion  picture  and  broadcast  applications. 

For  reproduction,  there  is  provided  a  new  light-weight  lateral  pick-up  having  high 
sensitivity  and  equipped  with  a  permanent  diamond  point.  This  reproducer,  in 
combination  with  its  associated  circuit,  is  suitable  for  use  on  all  lateral  cut  disk 
records. 

Pre-  and  post-equalization  are  employed  in  making  high-fidelity  records,  insuring 
a  low  noise  level.  This  reduction  of  background  noise  together  with  the  wide  frequency 
range  and  low  distortion  create  an  illusion  of  realism  or  "presence"  during  reproduc- 
tion. 

Usually  a  large  number  of  playings  is  not  required  from  direct  playback  disks. 
However,  because  of  the  low  mechanical  impedance  of  the  new  RCA  pick-up  and  the 
improved  composition  of  the  disks,  it  is  possible  to  reproduce  many  times  without 
appreciable  increase  in  noise  or  distortion. 

Constant  demand  for  better  direct  playback  recordings  has  resulted 
in  the  development  of  equipment  and  disks  of  improved  quality. 
These  advances  have  opened  new  fields  and  today  the  applications 
for  direct  playback  recording  are  numerous. 

The  art  has  advanced  well  past  an  experimental  state  and  is  now 
used  on  a  production  scale  for  sound  motion  pictures,  radio  broad- 
casting, schools,  industrial  advertising,  musical  education,  auditions, 
government  activities,  and  aviation. 

RCA  has  developed  direct  playback  equipment  having  improved 
performance  characteristics.  A  typical  studio  installation  is  shown 
in  Fig.  1.  Here  may  be  seen  the  feed-screw  mechanism  with  record- 
ing head  and  the  new  light-weight  lateral  reproducer. 

Lateral  or  "push-pull"  modulation  was  adopted  because  of  its  low 
distortion  characteristics.  A  frequency  range  of  50  to  10,000  cps.  is 

*Presented  at  the  1938  Fall  Meeting  at  Detroit,  Mich.;   received  October  24, 
1938. 

**RCA  Manufacturing  Co.,  Camden,  N.  J. 

246 


DISK  RECORDS  FOR  DIRECT  PLAYBACK 


247 


covered  with  reasonable  uniformity.  A  volume  range  of  approxi- 
mately 55  db.  is  obtainable  using  the  frequency  range  specified.  The 
records  have  sufficient  life  for  most  purposes  and  when  protected 
from  dust,  finger  prints,  and  oxidation  may  be  played  seventy-five 


FIG.  1. 


Recording  and  reproducing  system  for  broadcast 
studio  use. 


or  more  times  using  the  light-weight  flexible  pick-up  shown.  Usually 
a  large  number  of  playings  is  not  required  and  no  special  care  in  han- 
dling the  disks  is  ordinarily  demanded. 

The  MI  4887  recording  head  consists  of  a  mechanical  band-pass 
network  terminated  by  a  special  mechanical  resistance  material.    The 


DECIBELS 

III  4 

ft  o  u»  o  (j 

^^ 

^* 

•^ 

•  —  —  '      — 

~~ 

== 

•^ 

^+- 

^x 

^  — 

^ 

^ 

^^^^ 

*>* 

+*' 

0** 

FREQUENCY  IN  CYCLES  PER  SECOND 

FIG.  2.     Frequency  response  recorder  head  (MI-4887}  optical  measurement, 
lateral  stylus  velocity. 

impedance  of  the  moving  system  is  such  that  the  motion  of  the  cutting 
stylus  is  practically  unaffected  by  the  impedance  of  the  record  com- 
position. This  permits  a  free-space  microscopic  measurement  of 
the  performance  of  the  head,  which  is  duplicated  almost  exactly  when 
cutting  a  record.  The  frequency  response  characteristics  of  two  types 


248 


H.  J.  HASBROUCK 


[J.  S.  M.  p.  E. 


of  recorder  heads  are  shown  in  Fig.  2.  It  will  be  seen  that  frequen- 
cies below  800  cps.  are  controlled  so  as  to  hold  the  amplitudes  about 
constant,  the  stylus  velocity  diminishing  as  the  frequency  is  reduced. 


FIG.  3.     Construction]  of    recording  cutter   and   equivalent 
electrical  circuit. 

This  practice  is  followed  generally  in  disk  recording  to  avoid  cutting 
through  to  adjacent  grooves  at  low  frequencies.  The  loss  is  made  by 
suitable  compensation  in  the  reproducing  circuit. 

The  construction  of  the  recording  cutter  and  the  equivalent  elec- 
trical circuit  are  shown  in  Fig. 
3.  Since  for  constant  current 
in  the  recorder  winding,  the 
armature  receives  a  constant 
force,  the  electrical  circuit  is 
shown  working  from  a  con- 
stant-voltage source.  The  out- 
put of  the  mechanical  system 
is  represented  by  the  lateral 
stylus  velocity  and  is  equiva- 
lent to  current  through  the 
second  inductance  of  the  elec- 
trical network.  This  circuit  has 


/ 

^ 

/ 

/ 

x 

7 

s 

' 

- 

x-""^ 

INK 

=>EAK 
DRMAL 

LEVEL 
L*   USJ 

D 

MAX.  STYLUS    VELOCITY    INCHES    PER  SEC. 

FIG.  4.     Total  distortion  (rms.)  to  be 
expected  at  400  cps. 


a  rising  characteristic  with  increasing  frequency,  compensating  for 
diminishing  current  through  the  recorder  winding  due  to  inductance. 
The  merits  of  lateral  recording  have  been  discussed  at  length.    Of 
chief  importance  is  the  cancellation  of  even  harmonic  distortion.1 


Mar.,  1939] 


DISK  RECORDS  FOR  DIRECT  PLAYBACK 


249 


Under  normal  operating  conditions,  the  overall  distortion  of  the 
combined  recording  and  playback  operations  is  less  than  5  percent. 
The  rms.  total  distortion  to  be  expected  at  400  cps.  at  varying  de- 
grees of  modulation  is  indicated  in  Fig.  4.  These  observations  were 
made  at  a  record  speed  of  33.3  rpm.  and  a  diameter  of  12  inches. 


15          14          13          12          II          10  9  8 

DISC     DIAMETER     INCHES 

FIG.  5.     Average  losses  at  various  record  diameters  at 
33.3  rpm. 

All  disk  systems  suffer  from  high-frequency  transfer  losses  and  in- 
creased distortion  when  the  record  diameter  or  linear  surface  speed 
becomes  too  small.  The  average  losses  encountered  at  various  record 
diameters  at  33.3  rpm.  are  shown  in  Fig.  5.  These  are  for  soft  com- 
position blanks  used  in  direct  playback  work.  The  losses  are  less 
serious  on  standard  records  pressed  from  harder  material. 


</)   Ul 

Z   CD 


25-10 
a  z-15 


50 


100 


5000         10,000 


FIG.  6. 


500  1000 

FREQUENCY    CYCLES    PER    SECOND 
Reproducer  characteristic  with  standard  12-inch  pressing  at  33.3  rpm. 


These  losses  are  caused  by  a  combination  of  the  finite  size  of  the 
reproducing  point,  the  wavelength  of  the  recorded  sound,  the  weight 
of  the  pick-up,  and  the  compliance  of  the  record  stock.  It  is  possible 
to  compensate  to  some  extent  for  the  losses  but  it  has  not  been  found 
practicable  to  attempt  full  compensation  above  6000  to  7000  cps. 
because  of  the  serious  attenuation  in  the  upper  register.  Therefore, 
in  making  high-fidelity  records  where  quality  is  of  paramount  im- 
portance, the  best  results  are  obtained  by  dividing  the  time  into  two 
or  more  disks  and  confining  the  recording  to  reasonably  high  linear 
speeds. 


250 


H.  J.  HASBROUCK 


[J.  S.  M.  P.  E. 


Some  reduction  in  transfer  loss  can  be  effected  by  departing  from 
standard  groove  dimensions,  for  example,  by  reducing  the  radius 
from  0.0023  inch  to  0.001  inch.  A  further  reduction  could  be  made  by 
increasing  average  record  speed.  In  view  of  the  extensive  duplication 
of  equipment  neither  solution  appears  to  be  economical. 

The  MI-4856  reproducer  is  intended  primarily  for  use  on  non- 
abrasive  high-fidelity  records  but  may  be  used  on  all  lateral  records 
having  standard  groove  dimensions,  including  composition  coated 
disks  used  for  immediate  playback.  It  is  equipped  with  a  permanent 
diamond  point,  the  radius  of  which  corresponds  to  the  0.0023-inch 
standard.  This  radius  is  held  to  limits  not  exceeding  plus  or  minus 
0.0001  inch  to  insure  an  even  distribution  of  pressure  over  the  bottom 
of  a  standard  groove. 

The  frequency  response  characteristic  of  the  reproducer  from  a 
standard  test  pressing  twelve  inches  in  diameter  and  running  at  33.3 


-15 

50  IOO  500  1000  5000    10,000 

FREQUENCY    CYCLES    PER    SECOND 
FIG.  7.     Recording  and  reproducing  amplifier  charcteristics. 

rpm.  is  shown  in  Fig.  6.  The  internal  construction  of  the  pick-up  is 
shown  schematically  in  Fig.  8.  The  armature  is  of  the  clamped  reed 
type.  The  two  upper  air-gaps  are  inactive,  being  filled  by  non- 
magnetic spacers.  Since  the  armature  impedance  is  too  high  to  be 
directly  coupled  to  the  record  a  linkage  having  a  6  to  1  leverage  ratio 
or  a  36  to  1  impedance  ratio,  is  employed.  The  diamond  point  is 
secured  in  the  lower  end  of  an  extremely  light  pivot  arm  spring 
supported  vertically  but  rigid  laterally.  Thus,  the  pivot  arm  is  per- 
mitted to  rise,  as  during  "pinching,"  without  lifting  the  entire  pick- 
up. In  the  direction  of  useful  motion  being  transmitted  to  the  arma- 
ture, the  linkage  has  a  minimum  of  compliance  and  the  upper  cut-off 
is  high  or  about  9000  cps.  This  peak  is  reduced  by  means  of  a  block 
of  loaded  rubber  arranged  as  a  selective  damper  tuned  approximately 
to  the  peak  frequency. 

The  response  of  the  pick-up  working  into  a  resistance  load  would 
droop  at  high  frequencies  because  of  the  inductance  of  its  winding 


Mar.,  1939] 


DISK  RECORDS  FOR  DIRECT  PLAYBACK 


251 


unless  the  winding  were  kept  small.  This  is  not  consistent  with  high 
output.  Therefore  a  shunt  capacity  is  connected  across  the  pick-up 
which,  by  reacting  broadly  with  the  inductance,  increases  the  re- 
sponse through  a  portion  of  the  upper  range.  The  reproducer  has  a 
slightly  rising  characteristic  at  the  upper  end,  enough  to  offset  high- 
frequency  needle  or  transfer  losses  encountered  at  a  mean  record 
diameter  of  twelve  inches  at  33.3  rpm. 

In  making  high-fidelity  records,  including  disks  for  immediate  play- 
back, use  is  made  of  what  is  known  as  pre-  and  post-equalization  or 
complementary  compensation .  Because  of  the  energy  distribution  in 
most  speech  and  music,  it  is 
possible  to  accentuate  the  higher 
frequencies  when  making  a  record 
and  attenuate  them  in  reproduc- 
tion, thereby  reducing  the  sur- 
face or  ground-noise  due  to  the 
record  stock. 

Fig.  7  shows  the  recording  and 
reproducing  amplifier  characteris- 
tics and  the  ideal  overall  to  which 
must  be  added  the  characteris- 
tics of  the  reproducer.  This 
method  of  reducing  surface  noise 
can  be  used  successfully  in  most  cases  without  adding  appreciable 
distortion. 

Tests  indicate  negligible  wear  of  the  diamond  stylus  on  non- 
abrasive  records.  On  shellac  composition  records  there  is  sufficient 
wear  after  5000  ten-inch  faces  to  justify  replacement  of  the  point. 
This  is  considerably  longer  life  than  so-called  permanent  points  of 
iridium  or  sapphire,  when  used  on  abrasive  records  with  the  same 
pressure  of  two  ounces. 

An  improvement  in  pick-up  tracking  has  been  made  by  offsetting 
the  head  with  respect  to  the  arm.  This  angle,  which  is  about  ten 
degrees,  results  in  two  positions  of  tangency  with  the  groove,  one  near 
the  center  of  the  record  and  the  second  near  the  outer  edge.  The 
error  in  tracking  angle  between  these  positions  is  less  than  5  degrees. 

REFERENCE 


DIAMOND   /  \DRIVE   LINK 

FIG.  8.     Construction  of  pick-up. 


1  PIERCE,  J.  A.,  AND  HUNT,  F.  V.:  "Distortion  in  Sound  Reproduction  from 
Phonograph  Records,"  /.  Soc.  Mot.  Pict.  Eng.,  XXXI  (Aug.,  1938),  p.  157. 


252  H.  J.  HASBROUCK 

DISCUSSION 

MR.  WOLF  :     Is  this  a  conventional  form  of  acetate  ? 

MR.  HASBROUCK:  Yes.  The  response  was  equalized  to  9500  or  10,000  cps. 
There  was  no  limitation  below  that  either  in  recording  or  reproduction. 

MR.  WOLF:     Are  any  of  the  studios  now  making  use  of  it  for  playback? 

MR.  HASBROUCK:  Yes,  many  of  them,  both  motion  picture  and  broadcast 
studios. 

MR.  CARVER:  When  you  say  the  "conventional  type  acetate"  I  suppose  you 
mean  nitrate.  All  I  have  ever  seen  was  nitrate. 

MR.  HASBROUCK:  That  word  "acetate"  is  misused.  It  is  a  nitrate  base. 
There  are  other  ingredients. 

MR.  CRABTREE  :  What  are  the  merits  of  this  lateral  system  as  against  the  hill- 
and-dale? 

MR.  HASBROUCK:  That  has  been  discussed  at  great  length  in  many  publica- 
tions. The  chief  advantage  is  the  cancellation  of  even-harmonic  distortion  in  the 
lateral  system.  It  is  similar  to  a  "push-pull"  amplifier. 

MR.  JOHNSON:  The  hill-and-dale  controversy  is  like  an  automobile  running 
over  a  rough  road,  compared  to  an  engine  on  a  wavy  track.  The  best  contact 
between  the  needle  and  the  groove  will,  of  course,  be  in  the  lateral  cut  track. 
When  the  stylus  digs  deeper  the  drag  is  greater,  and  there  is  likely  to  be  chattering. 

MR.  CRABTREE  :  Many  years  ago  we  had  demonstrations  at  these  meetings  of 
hill-and-dale  records  which,  to  me,  were  quite  pleasing.  It  is  unfortunate  that  the 
author  did  not  bring  along  his  recorder  and  actually  record  in  the  room  and 
reproduce,  say,  a  speech  by  our  President.  I  wish  we  could  arrange  at  the  next 
meeting  for  someone  to  put  on  such  a  demonstration.  It  means  so  much  more 
when  we  hear  the  original  and  then  the  reproduction,  than  to  hear  merely  the 
reproduction. 

MR.  WOLF:  Is  this  system  used  in  Hollywood  to  the  exclusion  of  other  play- 
back methods?  Is  the  Miller  system  being  used  at  all  for  immediate  playback: 

MR.  WOLFE  :  To  the  best  of  my  knowledge,  the  Miller  system  is  not  being  used 
in  the  studios,  but  this  method  is  certainly  not  the  only  one  in  use  for  playback 
purposes. 

Again,  so  far  as  I  know,  variable-density  negative  is  not  being  played  back  in 
the  studios,  but  R.K.O.  does  play  back  variable-area  negative  in  certain  cases. 
Two  methods  of  playing  back  are  used :  either  acetate  disk,  some  of  it  lateral  and 
some  hill-and-dale,  or  by  reproduction  from  film. 

MR.  WOLF:  I  think  before  long  electromagnetic  recording  will  probably  be  in 
the  studios.  The  Bell  Laboratory  has  been  working  for  a  number  of  years  on  that 
method  of  recording.  I  think  you  will  see  at  the  New  York  World's  Fair,  if  not 
before,  a  quality  of  magnetic  recording  comparable  with  the  best  playbacks  most 
of  us  have  ever  heard.  That  medium  will  also  be  very  useful  for  certain  kinds  of 
playback  and  editing. 


THE  CENTENARY  OF  PHOTOGRAPHY  AND  THE 
MOTION  PICTURE* 

EDWARD  EPSTEAN** 


January  7,  1939,  marked  the  centenary  of  the  day  when  Arago  communicated  to 
the  French  Academy  of  Sciences  news  of  the  invention  by  Daguerre  of  what  was  to  be 
known  as  the  daguerreotype. 

At  the  suggestion  of  the  Societe  Fran$aise  de  Photographie  that  ceremonies  and  meet- 
ings be  held  by  scientific  and  engineering  bodies  throughout  the  world  to  commemorate 
this  Centenary,  the  Atlantic  Coast  Section  of  the  SMPE  devoted  its  January,  1939, 
meeting  to  two  presentations:  the  one  that  follows,  by  Edward  Epstean,  describing 
the  historical  background  of  Daguerre' s  time,  as  related  to  photography,  or  "light 
writing"]  and  the  second  by  Beaumont  Newhall,  describing  more  in  detail  the  work 
of  Daguerre  and  his  process. 

The  erroneous  popular  idea  that  engineers  are  mechanics,  with  no 
interest  in  the  history  or  philosophy  of  their  profession  is,  I  am  sure, 
not  held  by  the  members  of  this  Society.  I  do  not  hesitate,  therefore, 
on  this  Centenary  of  the  "Discovery  of  Photography,"  to  address 
you  on  this  subject,  tracing  the  history  and  progress  of  the  art  lead- 
ing to  your  specialized  science  of  motion  picture  engineering. 

Motion  pictures,  as  we  know  them  today,  were  originally  called 
"animated  photographs"  and  later  "moving  pictures." 

Robert  Hunt,  the  English  professor  of  physical  science,  wrote  in 
1854 :  "The  progress  of  discovery  is  ordinarily  slow,  and  it  often  hap- 
pens that  a  great  fact  is  allowed  to  lie  dormant  for  years,  or  for  ages, 
which,  when  eventually  revived,  is  found  to  render  a  fine  interpreta- 
tion of  some  of  Nature's  harmonious  phenomena  and  to  minister  to 
the  wants  or  the  pleasures  of  existence.  Of  this  position,  Photography 
is  peculiarly  illustrative." 

The  universal  application  of  photography,  which  ranges  from  the 
hobby  of  the  amateur  on  through  the  wide  fields  of  science  and  in- 
dustry, might  suffice  to  arouse  in  any  one  the  desire  for  a  short  study 
of  its  antecedents  and  its  early  development.  This  year  is  extremely 


*Presented  at  a  meeting  of  the  Atlantic  Coast  Section,  January  11,  1939. 
**  New  York.  N.  Y. 


253 


254  E.  EPSTEAN  [j.  s.  M.  p.  E. 

appropriate,  because  January  7,  1939,  marks  the  Centenary  of  the 
day  when  the  French  scientist,  Arago,  member  of  the  Chamber  of 
Deputies  and  of  the  Academy  of  Sciences,  communicated  the  first 
news  of  this  new  discovery  to  the  members  of  the  Academy.  After 
a  few  months  of  necessary  political  procedure,  the  French  Govern- 
ment, through  Arago,  published  the  details  of  the  process — not  of 
photography,  but  of  the  daguerreotype. 

To  give  a  full  history  of  what  today  is  called  photography  with  its 
background,  development,  and  innumerable  applications  would  re- 
quire many  volumes.  The  word  photography,  from  its  derivation, 
implies  primarily  a  study  of  light,  and  early  in  Genesis  the  word  is 
mentioned  in  its  very  first  verses.  Ever  since,  philosophy  and  science 
have  attempted  without  success  to  define  and  explain  Light.  Dr. 
Woodbridge,  of  Columbia  University,  writes  in  his  chapter  on  the 
subject  that:  "it  is  a  paradox"  and  it  probably  will  remain  so,  in 
saecula  saeculorum.  We  know  little  of  its  action  and  you  will  easily 
understand  my  meaning  when  I  call  your  attention  to  the  use  of  the 
phrase:  "seeing  the  sun,  the  moon,  the  stars."  What  we  see,  of 
course,  is  only  the  radiation  of  their  light  and  not  its  source.  Dis- 
tance, time  and  space,  its  brilliancy,  and  its  movement,  our  imperfect 
optical  apparatus — all  make  it  impossible  to  see  that  essence  which 
we  include  in  the  term  "the  light."  I  have  read  that  stars  which  have 
been  extinct  for  400  years  still  impress  our  vision  with  their  light. 

Resuming  our  study  of  photography  we  find  the  camera  obscura. 
Some  historians  have  tried  to  trace  its  origin  to  the  Arabs.  Astrologers 
were  attached  to  the  courts  of  their  rulers,  whose  lives,  fortunes,  and 
wars  were  influenced  by  the  astrological  studies  of  the  planets,  based 
on  the  day  and  hour  of  the  ruler's  birth.  In  order  to  study  them 
apart  from  the  myriads  of  other  stars  surrounding  them,  the  seers 
built  observation  huts  where  no  light  could  enter  save  through  a  hole 
in  the  roof  of  the  "dark  room." 

Leonardo  da  Vinci  in  his  notes  on  optics  states :  "...  if  the  front  of 
a  building  .  .  .  which  is  illuminated  by  the  sun  has  a  dwelling  over 
against  it,  and  in  that  part  of  the  front  which  does  not  face  the  sun 
you  make  a  small  round  hole,  all  the  objects  which  are  lighted  by  the 
sun  will  transmit  their  images  through  this  hole,  and  will  be  visible  in- 
side the  dwelling  on  the  opposite  wall  which  should  be  made  white. 
And  they  will  be  there  exactly  but  inverted;  and  if  in  different  parts 
of  the  same  wall  you  make  similar  holes  you  will  produce  the  same 
effect  in  each."  And  thus  we  have  "light"  writing  through  a  so-called 


Mar.,  1939]  CENTENARY  OF  PHOTOGRAPHY  255 

"pinhole  lens"  on  a  white  wall  or  on  white  paper  fastened  opposite 
the  hole. 

Cameras,  front  boards,  bellows,  and  plate  holders  have  since  been 
constantly  improved.  From  the  beginnings  of  photography  the  cam- 
era obscura  was  equipped  with  a  single  lens,  but  soon  thereafter  came 
the  periscopic  lens,  the  meniscus  prism,  and  double  objectives. 

Astrologers  were  displaced,  largely  through  the  greed  for  precious 
metals,  by  the  alchemists,  forerunners  of  the  chemist — in  our  special- 
ized field  the  photochemist. 

Northern  Italy,  which  provided  colored  ribbons  and  fabrics  for 
Europe  during  the  eighteenth  century,  was  naturally  interested  in 
the  action  of  dyes  and  colors,  and  to  Beccarius  is  ascribed  the  priority 
of  discovery  of  the  light  sensitivity  of  chloride  of  silver.  An  earlier 
work,  in  Latin,  by  Agricola  (1490-1555)  deals  with  silver  ores.  This 
work  was  translated  into  English  by  President  and  Mrs.  Hoover. 
Here  is  one  of  the  examples  of  a  comatose  condition,  because  a  closer 
study  of  this  work  might  have  hastened  an  understanding  of  the 
action  of  light  on  the  silver  salts.  It  is  quite  certain  that  the  al- 
chemists of  the  sixteenth  century  knew  of  the  aqueous  production  of 
silver  chloride. 

Again  we  see  a  dormant  period  in  the  progress  of  the  discovery  of 
photography,  which  extends  to  the  end  of  the  eighteenth  century.  One 
of  the  men  to  whom  great  credit  is  due  in  this  celebration,  dis- 
covered— again  not  photography  but  as  he  named  it — heliography. 
Joseph  Nicephore  Niepce  (1765-1833),  educated  for  the  priesthood, 
therefore  equipped  with  a  knowledge  of  the  humanities  and  a  training 
in  elementary  science,  was  drawn  into  the  vortex  of  the  wars  of  the 
French  Revolution.  He  was  discharged  from  the  army  at  his  request, 
being  debilitated  by  a  fever  which  at  the  time  raged  both  in  the 
army  and  among  the  civilian  population. 

Niepce's  part  in  the  invention  of  photography  is  best  stated  in  his 
own  words.  He  described  his  process  as  "automatically  fixing  by 
the  action  of  light  the  image  formed  in  the  camera  obscura  and  the 
reproduction  by  printing  with  the  aid  of  known  processes  of  engrav- 
ing." Niepce  experimenting  with  lithography  (1813-1815),  then 
new  in  the  reproductive  process,  attempted  to  obtain  designs  on  stone 
and  metal  by  the  action  of  light  instead  of  by  manual  drawing,  and 
thus  produced  etched  intaglio  plates.  He  used  diaphragms  in  his  lens 
and  added  bellows  to  his  camera.  His  first  camera  images  were  ob- 
tained in  1816,  at  the  end  of  which  year,  being  unable  to  fix  his  paper 


256  E.  EPSTEAN  [J.  S.  M.  P.  E. 

negatives,  he  abandoned  the  use  of  silver  chloride  and  began  experi- 
ments with  asphaltum.  While  successful  in  the  copying  of  engravings 
by  contact,  he  turned  back  in  1826,  substituting  glass  for  metal  and 
paper,  to  intaglio  etching  of  images  obtained  in  the  camera.  But  he 
did  not  succeed  in  reproducing  the  middle  tones,  and  in  that  year — 
1826 — Daguerre  heard  of  Niepce  through  the  Paris  optician  Chevalier, 
and  in  1829  we  find  Niepce  sending  to  Daguerre  "a  view  from  nature 
engraved  on  a  silvered  pewter  plate."  He  met  Daguerre  personally 
in  August,  1827,  on  a  trip  through  Paris.  At  that  time  Daguerre's 
progress  had  resulted  in  nothing  but  fantastic  experiments,  without 
any  significance  in  obtaining  images  by  the  action  of  light,  while 
Niepce  had  achieved  actual  results  before  the  end  of  1822.  Toward 
the  end  of  1827,  while  visiting  his  dying  brother  in  England,  Niepce 
presented  to  the  Royal  Society  a  short  memoir  entitled  "Heliog- 
raphie,  dessins  et  gravures."  Since,  however,  he  did  not  disclose  the 
details  of  his  manipulations  the  communication  was  returned,  un- 
read, by  the  Society.  Returning  to  Paris  in  January,  1828  he  was 
urged  by  his  friends  during  his  stay,  which  was  prolonged  until  the 
end  of  February,  to  join  Daguerre  in  perfecting  and  exploiting  his 
invention.  It  was  in  1829  that  Niepce  first  used  iodine  to  blacken 
the  silvered  background  of  his  images.  In  December  of  that  year 
articles  of  partnership  were  drawn  at  Chalon-sur-Sa6ne  between 
Niepce  and  Daguerre,  on  a  visit  Daguerre  paid  to  Niepce  for  this 
purpose.  Niepce  died  on  July  5,  1833,  in  his  sixty-ninth  year. 

Louis  Mande  Daguerre  (1767-1851),  the  other  pioneer  honored  at 
this  Centenary,  had  none  of  the  culture  and  scientific  training  which 
Niepce  enjoyed  in  his  youth.  He  was  preeminently  an  artist,  blessed 
with  imagination,  seeking  fame  and  publicity,  and  fortunate  in  having 
powerful  friends.  We  have  no  record  of  his  having  had  any  prepara- 
tion for  chemistry  or  optics  of  photography.  But  he  showed  a  positive 
genius  for  adapting  the  ideas  of  others.  His  share  in  the  invention  of 
photography  was  not  disclosed  until  six  years  after  the  death  of  his 
partner,  to  whom  little  credit  was  given  in  the  publication  of  the  da- 
guerreotype process.  Having  been  sent  to  Paris  in  his  youth  to  study 
art,  he  eventually,  in  collaboration  with  the  distinguished  painter, 
Bouton,  created  the  Diorama,  a  marked  improvement  on  the  pano- 
rama. Incidentally,  the  panorama  was  introduced  in  Paris  by  the 
American,  Robert  Fulton,  the  inventor  of  the  steamboat,  during  a 
visit  from  1800-1804.  Daguerre's  novel  lighting  effects  added  to 
the  mobility  of  the  scene  and  to  the  attractiveness  of  the  colors  in  the 


Mar.,  1939]  CENTENARY  OF  PHOTOGRAPHY  257 

views  displayed.  Daguerre  was  made  a  Chevalier  of  the  Legion  of 
Honour  in  1824  and  it  is  said  that  the  profits  of  the  Diorama  in  the 
same  year  amounted  to  two  hundred  thousand  francs.  It  is  this 
financial  success  which  permitted  him  in  his  leisure  time  to  study 
and  to  make  the  experiments  for  finding  the  means  of  fixing  the 
image  obtained  in  the  camera  obscura.  The  Diorama  burned  on 
March  3,  1839,  inflicting  a  serious  financial  loss  on  Daguerre.  The 
fact  is  indisputable  that  Daguerre  was  the  inventor  of  the  daguerreo- 
type, and  one  of  the  inventors  of  photography.  He  must  be  given 
credit,  as  the  first  to  recognize  the  light  sensitivity  of  iodide  of  silver 
as  well  as  the  property  of  vapors  of  mercury  to  reveal  the  latent  image. 
Whether  he  discovered  these  things  by  chance  or  whether  he  built 
on  the  pioneer  knowledge  of  Niepce  and  others  before  him  can  not 
rob  him  of  this  honor.*  He  died  on  July  10,  1851,  in  his  eighty- 
fourth  year. 

The  earliest  information  given  to  the  public  about  the  daguerreo- 
type was  the  report  made  by  Arago  to  the  Academy  of  Sciences  on 
January  7,  1839,  as  chairman  of  the  committee,  consisting  of  Hum- 
boldt,  Biot,  and  himself,  appointed  to  visit  Daguerre.  His  speech 
before  the  Chamber  of  Deputies  on  July  3,  1839,  was  followed 
by  a  similar  address  by  Gay-Lussac  in  the  Chamber  of  Peers  on  July 
30th,  in  support  of  the  bill  acquiring  the  purchase  of  the  daguerreo- 
type and  diorama  by  the  French  government.  A  few  days  later  the 
Minister  of  the  Interior  instructed  Arago  to  promulgate  the  processes 
to  the  world,  and  this  Arago  did  in  his  address  to  the  Academy  of 
Sciences  on  August  19,  1839.  This  splendid  gesture  of  the  French 
government  in  freely  giving  this  portentous  discovery  to  the  world 
can  not  be  emphasized  sufficiently. 

Arago's  famous  address  to  the  Academy  contained  no  technical  de- 
tails. These  were  later  printed  in  the  official  handbook  of  daguerreo- 
typy:  Historique  et  Description  des  Precedes  du  Daguerreotype 
et  du  Diorama,  written  by  Daguerre,  which  passed  through  several 
editions  and  appeared  in  most  foreign  languages  before  the  end  of 
the  year.  No  previous  discovery  in  history  had  awakened  such  uni- 
versal interest. 

*  Briefly,  the  Daguerre  process  is  as  follows :  A  metal  plate  is  thoroughly  cleaned 
and  polished  and  is  then  coated  with  silver,  which  is,  in  turn,  highly  polished. 
Metallic  iodine  is  then  sublimed  on  the  silver  coating,  producing  the  light-sensi- 
tive silver  iodide.  After  exposure  the  image  is  developed  by  fumes  of  mercury 
and  is  fixed  with  sodium  thiosulfate,  or  hypo. 


258  E.  EPSTEAN  [J.  s.  M.  P.  E. 

Into  this  chorus  of  universal  praise,  however,  there  came  a  dis- 
cordant strain  from  England.  Daguerre,  in  England  alone,  had  pat- 
ented his  invention.  Talbot,  the  English  scientist,  claimed  priority 
for  the  invention  of  his  process  of  "photogenic  drawing,"  which  he 
communicated  to  the  Royal  Society  on  January  31,  1839. 

Let  us  look  into  the  history  of  the  beg  nnings  and  development  of 
photography  in  England. 

Thomas  Wedgwood,  during  the  years  from  1792  to  1800,  carried  on 
experiments  which  he  published  in  the  Journal  of  the  Royal  Institu- 
tion, London,  under  the  title  "An  Account  of  the  Method  of  Copying 
Paintings  upon  Glass,  and  of  Making  Profiles,  by  the  Agency  of 
Light  upon  Nitrate  of  Silver,  Invented  by  T.  Wedgwood,  Esq.,  with 
Observations  by  H.  (Sir  Humphry)  Davy."  Wedgwood  was  the  first 
to  conceive  the  idea  of  delineating  the  form  of  objects  by  the  action 
of  light,  but  he  died  without  having  found  the  means  of  fixing  (mak- 
ing permanent)  these  photographic  images. 

Sir  John  (John  Frederick  William)  Herschel  (1792-1871),  a  great 
scientist,  who  knew  and  applied  the  axiom  that  science  without 
philosophy  was  like  the  body  without  mind,  directed  his  researches 
to  the  physical  laws  governing  chemical  reactions.  It  is  he  who 
pointed  out  that  hyposulfite  of  soda  is  the  best  fixing  agent. 

After  careful  research,  I  do  not  hesitate  to  state  that  it  was  Sir 
John  Herschel  who  coined  the  word  "photography."  Dr.  Erich 
Stenger's  claim  that  it  was  the  German  astronomer  Madler  who  first 
used  the  term  (February  25,  1839)  was  dispelled  in  my  mind  at  the 
time  of  a  memorable  visit  which  I  paid  to  Miss  Hardcastle,  Sir  John's 
granddaughter,  who  resides  in  Observatory  House  at  Slough,  near 
London — the  home  of  Sir  John.  I  found  there  considerable  cor- 
respondence between  the  two  astronomers,  and  I  have  no  doubt  that 
a  close  search  would  disclose  the  use  of  the  word  in  Sir  John's 
correspondence  with  Dr.  Madler  before  the  publication  in  the  news- 
paper column  cited  by  Professor  Stenger. 

William  Henry  Fox  Talbot  (1800-1877),  philologist  and  arche- 
ologist,  states:  "In  the  spring  of  1834  I  began  to  put  in  practice  a 
method  of  employing  .  .  .  the  property  .  .  .  possessed  by  nitrate  of 
silver  ...  its  discoloration  when  exposed  to  the  violet  rays  of  light." 
He  published  his  "experiments"  under  the  name  "Photogenic  Draw- 
ing" in  1839,  explaining  the  preparation  of  photogenic  paper,  the 
washing,  drying,  and  coating  of  it  and  his  success  in  increasing  its 
sensitivity — exposure  of  five  minutes.  It  was  early  in  1840  that  the 


Mar.,  1939]  CENTENARY  OF  PHOTOGRAPHY  259 

Calotype  process  (kallos  meaning  beautiful  in  Greek)  reached  the  con- 
tinent but  daguerreotypes  were  preferred,  at  least  until  Blanquart- 
Evrard  perfected  Talbot's  process.  The  great  advantage  of  the 
Talbotype  was,  of  course,  its  ability  to  multiply  the  record  obtained 
in  the  negative  by  any  number  of  positive  prints. 

We  must  not  pass  over  this  period  without  mentioning  the  Scot, 
Mungo  Ponton,  who  laid  the  foundation  for  the  present  relief  and 
other  reproduction  processes  by  announcing  the  light-sensitive  prop- 
erty of  bichromate  of  potash. 

During  all  this  time  improvements  in  the  design  and  construction 
of  cameras  and  lenses  were  introduced,  new  researches  in  photo- 
chemistry and  innumerable  methods  of  making  photographic  copies. 

A  short  American  note  may  be  given  place  here.  In  no  other 
country  in  the  world  was  daguerreotypy  more  enthusiastically  re- 
ceived or  so  widely  practiced  as  in  America.  A  very  complete  account 
of  the  early  American  procedure  in  daguerreotypy,  by  Dr.  J.  W. 
Draper,  of  New  York  was  published  in  the  London  and  Edinburgh 
Philosophical  Magazine  for  September,  1840. 

Daguerreotypes  and  Calotypes,  however,  were  shortly  to  be  dis- 
placed by  the  wet  collodion  process.  Domont  and  M£nard  recognized 
in  1847  the  solubility  of  certain  kinds  of  gun  cotton  in  ether  and  al- 
cohol, giving  us  collodion,  the  name  of  which  is  derived  from  the 
Greek  word  denoting  "like  glue — sticky."  The  credit  for  its  first 
use  in  photography,  January,  1850,  is  ascribed  by  Dr.  Eder  to  Gustave 
le  Gray  in  Paris.  A  closer  study  of  the  subject,  however,  confers  the 
merit  of  introducing  the  first  practical  collodion  process  on  the 
Englishman,  Frederick  Scott  Archer.  It  displaced  all  previous 
methods  in  the  decade  preceding  1860,  and  for  twenty  years  the  wet 
collodion  process  occupied  the  first  place  among  photographic  nega- 
tive methods.  In  the  seventies  along  came  the  bromo-silver  dry 
collodion  emulsion  with  excess  of  silver  nitrate,  followed,  late  in  that 
decade,  by  gelatine  bromide-silver  emulsions,  alkaline  development — 
the  gelatine  dry  plate  era.  Here  we  meet  George  Eastman,  introduc- 
ing the  stripping  film  and  roll  holder,  with  later  the  Kodak  camera 
and  daylight-loading  roll-film — modern  photography. 

Chronology  now  leads  us  to  interject  here  brief  mention  of  stereo- 
scopic photography.  The  principle  of  binocular  vision  was  known 
in  remote  times  but  just  a  hundred  years  ago,  the  English  physicist, 
Charles  Wheatstone,  of  Gloucester,  England  (1802-1875),  invented 
(1838)  the  mirror  stereoscope,  through  which  one  viewed  two  slightly 


260  E.  EPSTEAN  [j.  s.  M.  P.  E. 

dissimilar  images  of  the  same  object,  as  seen  by  the  two  eyes,  and  ob- 
tained a  single  image  having  the  natural  solidity  or  relief  of  the  object. 
In  Wheatstone's  instrument,  of  course,  only  geometrical  designs 
could  be  used.  Figures  and  scenes  were  unattainable  until  pho- 
tography supplied  the  means  by  making  two  simultaneous  exposures 
of  the  same  subject  with  two  lenses  placed  equally  apart  from  a 
median  line. 

Sir  David  Brewster  (1781-1868)  replaced  Wheatstone's  mirrors 
with  round  prisms  and  in  1844  produced  a  practical  apparatus,  which 
we  know  as  the  refracting  or  lenticular  stereoscope.  And  it  is  here  that 
we  find  the  first  liaison  with  animated  photography. 

The  chrysalis  from  which  the  metabulous,  beautiful,  brightly 
colored  butterfly — our  cinema — emerged,  seems  to  have  been  the 
stroboscope — strobos  is  derived  from  a  Greek  word  meaning  "a  whirl- 
ing." But  a  thousand  years  before  1833,  the  year  when,  within  a  few 
months  of  each  other,  the  Belgian  Plateau  and  the  Austrian  Stampfer 
developed  their  inventions,  a  Roman  poet,  Lucretius  Careus,  wrote 
a  verse  practically  dealing  with  "moving  pictures."  Translated  from 
the  Latin  it  is  something  as  follows : 

"Verily,  it  is  no  miracle  when  images  move  their  arms  and  other 
members  of  their  bodies  around,  in  rhythmic  time.  Indeed,  as  one 
passes  another  appears  in  different  pose,  the  former  seems  to  have 
changed  its  gestures.  The  change  of  course  must  take  place  rapidly." 

However,  Lucretius  expresses  only  the  fact  of  the  persistence  of 
vision,  not  the  mechanics  or  the  demonstration. 

Wilfred  Funk  in  his  book  "So  You  Think  It's  New,"  attempts  to 
go  still  farther  back  and  writes:  "Movies?  They  had  them  in  far-off 
Greece.  Pictures  were  painted  on  pillars  in  progressive  fashion,  the 
idea  being  to  ride  by  them  on  horseback  and  thus  get  the  effect  of 
motion.  Then  some  smart  inventor  devised  a  better  method.  He 
painted  a  series  of  pictures  in  spiral  sequence  on  a  single  revolving 
pillar.  This  was  spun  by  a  rope  and  thus  the  audience  enjoyed  the 
first  cinema." 

Plateau  and  Stampfer  painted  moving  figures,  for  instance,  that  of 
a  dancer,  on  the  periphery  of  a  disk  and  on  that  of  a  cardboard  they 
cut  slits,  the  number  of  which  depended  on  the  series  of  subjects 
and  the  speed  of  the  motion.  These,  turned  quickly  by  hand,  re- 
volved on  their  axes  in  front  of  a  mirror,  producing  the  illusion  of 
movement  in  the  figures. 


Mar.,  1939]  CENTENARY  OF  PHOTOGRAPHY  261 

Faraday  took  up  the  subject  of  dividing  the  apparent  animate 
movement  of  objects  from  the  inanimate  and  vice  versa.  He  read  a 
paper  before  the  Royal  Society  in  1831  on  "A  Peculiar  Class  of 
Optical  Deceptions  Showing  Wheel  Phenomena." 

Plateau  called  his  apparatus  phenakistiscope,  phantasmascope,  and 
phantascope.  In  Austria  they  were  called  zoetropes  and  other  names. 
In  1833-1834  the  Englishman  Homer  described  the  "marvel  drum." 
The  Scotch  physicist  Maxwell  constructed  such  a  drum  with  optical 
adjustment  by  inserting  concave  lenses  into  the  viewing  slits  of  the 
drum.  The  American  patent  of  A.  B.  Brown,  August  10,  1839,  is  the 
first  to  mention  the  rapidly  changing  exposure,  interrupted  with  the 
aid  of  a  rotating  shutter  and  at  the  same  time  simultaneously  inter- 
rupting the  picture  plate.  This  arrangement  corresponded  to  the 
Maltese  Cross  of  modern  motion  picture  apparatus  and  presents  in 
its  essential  characteristics  the  modern  motion  picture  machine. 

I  have  no  doubt  you  are  better  informed  than  I  am  on  the  develop- 
ment of  the  motion  picture  industry  in  the  photographic  field,  its 
cameras  and  accessories,  its  reproduction  of  color,  and  its  synchroni- 
zation with  sound.  At  any  rate,  the  technical  journals  specializing 
in  motion  picture  engineering  and  production  give  a  much  better 
presentation  of  the  subject  than  I  could  possible  do.  However,  my 
cursory  study  of  the  subject  and  the  information  gained  from  the 
Historical  Committee  of  your  Society  have  given  me  certain  informa- 
tion which  I  have  combined  in  the  following  paragraphs. 

Edward  Muybridge,  born  in  England  in  1830,  emigrating  to  Amer- 
ica, became  a  professional  photographer  and  was  appointed  direc- 
tor of  Photographic  Survey  of  the  California  Coast.  In  1872  he  at- 
tracted the  attention  of  Governor  Leland  Stanford  of  California,  who 
was  a  lover  of  horses  and  kept  a  racing  stable.  A  wager  on  the  ques- 
tion of  whether  a  horse  when  running  at  full  speed  touched  the  ground 
with  one  or  more  feet,  led  Governor  Stanford  to  the  desire  of  register- 
ing this  motion.  At  about  the  same  time  a  Frenchman,  Jules  Marey, 
pursuing  an  investigation  along  the  same  lines,  had  devised  a  system 
for  the  analysis  of  movement,  which  he  called  Chronography.  Gover- 
nor Stanford,  who  heard  of  this,  had  the  happy  idea  to  settle 
the  wager  by  the  aid  of  photography,  an  idea  quite  novel  when 
we  consider  photographic  technic  at  that  time.  Marey  had  never 
thought  of  such  a  possibility  but  Muybridge  practically  devoted  his 
whole  life  to  the  perfecting  of  motion  photography.  In  the  analysis 
of  movement  Marey  and  Muybridge  kept  each  other  informed  of  the 


262  E.  EPSTEAN  [J.  s.  M.  P.  E. 

progress  of  their  work  and  the  experiments  of  these  two  men  make 
an  interesting  study. 

Researches  by  Terry  Ramsaye,  reported  in  "A  Million  and  One 
Nights — the  History  of  the  Motion  Picture,"  indicated  that  little  or 
no  progress  was  made  in  the  solution  of  the  problem  by  Muybridge 
and  that  Governor  Stanford  finally  called  in  an  engineer,  John  D. 
Isaacs,  who  worked  out  the  fundamentals  of  the  scheme,  tried  out 
the  scheme  successfully,  and  then  turned  it  over  to  Muybridge.  Dur- 
ing the  succeeding  years,  Muybridge  applied  these  ideas  with  con- 
siderable skill,  but  did  not  show  any  originality  in  improving  upon 
them.  The  date  when  Isaacs  obtained  satisfactory  pictures  by  this 
method  was  about  1882,  although  Muybridge  has  often  been  er- 
roneously credited  with  working  out  the  scheme  several  years  prior 
to  that  date.  He  died  in  1904. 

Ducos  du  Hauron  patented  in  1864:  "An  apparatus  having  for  its 
purpose  the  photographic  reproduction,  in  any  quantity,  of  a  scene 
with  all  the  changes  to  which  it  is  subjected  during  a  specified  time." 
The  apparatus  was  never  constructed.  The  details  of  the  patent 
were  never  published  and  his  methods  are  unknown  today.  He 
was  far  ahead  of  his  time.  His  arrangement  of  rotating  lenses  was 
realized  thirty  years  later  by  the  American  Jenkins. 

C.  Francis  Jenkins  (1867-1934)  played  a  noteworthy  role  in  con- 
nection with  motion  picture  photography  in  America.  Jenkins  was 
the  founder  of  the  Society  of  Motion  Picture  Engineers.  He  con- 
structed in  1893  a  motion  picture  camera  called  the  "Phantoskop" 
which  was  described  in  the  "The  Photographic  Times,"  Vol.  25,  p.  2, 
July  6,  1894.  A  peep-hole  type  of  viewing  machine  was  also  invented 
by  Jenkins  and  a  patent,  U.  S.  No.  536,539  was  issued  to  him  on 
March  26,  1895. 

Research  by  your  Historical  Committee  has  shown  that  these  early 
devices  of  Jenkins  had  no  commercial  practicability,  and  that  it  is 
Thomas  Armat,  rather  than  Jenkins,  who  is  entitled  to  major  credit 
for  the  design  of  the  first  successful  motion  picture  projector.  In 
subsequent  years,  Jenkins  devised  and  developed  valuable  inven- 
tions related  to  the  science  of  motion  picture  engineering,  among 
which  the  high-speed  camera  was  one  of  the  most  successful. 

The  contributions  of  William  Friese-Greene,  an  English  photog- 
rapher, and  those  of  our  own  Thomas  Alva  Edison  (1847-1931)  fall 
within  our  own  times  and  within  the  much  more  comprehensive  com- 
pass of  your  own  experience. 


Mar.,  1939]  CENTENARY  OF  PHOTOGRAPHY  263 

Auguste  and  Louis  Lumi£re  of  Lyons,  France,  not  only  coined  the 
word  "cine'matographe"  and  gave  their  first  exhibition  at  Paris 
in  1895,  but  were  also  the  first  to  place  on  the  market  remarkably 
simple  and  efficient  apparatus  for  taking  and  projecting  serial  pic- 
tures in  which  the  perforated  film  strip  was  for  the  first  time  held 
and  moved  by  a  gripper.  It  is,  of  course,  well  known  that  this  firm 
had  been  making  dry  plates  for  years.  The  history  of  the  Brothers 
Lumiere  and  their  accomplishments  deserve  a  special  biography, 
for  their  services  to  photography  in  all  its  branches  are  extremely  im- 
portant. 

In  conclusion,  I  can  but  ask  you  to  compare  your  present  palatial 
and  itinerant  motion  picture  camera  on  automobile  trucks  with  the 
apparatus  with  which  the  early  American  photographers  had  to  con- 
tend. Professor  Robert  Taft,  in  his  splendid  book  on  Photography  and 
the  American  Scene  recently  published  by  Macmillan,  speaks  of  "in- 
dependent photographers  who  were  gradually  pushing  west  (or  east 
from  the  west  coast)  as  the  line  of  the  frontier  gradually  changed." 
He  speaks  of  C.  E.  Watkins  of  San  Francisco,  who  was  born  in 
New  York  and  went  to  California  as  a  young  man  around  the 
1850's.  In  1861  ''Watkins  made  his  first  trip  into  the  Yosemite 
Valley,  which  was  disinguished  by  the  fact  that  he  took  with  him  a 
camera  constructed  by  himself,  capable  of  taking  a  plate  18  X  22 
inches  in  size.  The  use  of  these  large  plates  by  a  wet  plate  photog- 
rapher, working  under  the  most  favorable  circumstances,  was  at- 
tended with  considerable  difficulty.  It  was  a  feat  of  no  mean  skill 
to  flow  the  collodion  on  these  plates,  obtain  a  uniform  film,  and  then 
sensitize,  expose,  and  develop  them  when  they  were  still  moist.  .  .  A 
twelve-mule  train  was  necessary  to  transport  Watkins  and  his  sup- 
plies to  the  valley,  and  five  of  these  mules  carried  his  equipment  as 
he  made  his  photographic  tour  of  the  region.  As  each  photograph 
was  made  the  darkroom  tent  had  to  be  unpacked  and  set  up,  the 
plates  prepared  in  the  small  tent,  and  developed  immediately  after  ex- 
posure. The  equipment  was  then  repacked  and  the  mule  train  moved 
on  to  obtain  the  next  view." 

The  daguerreotype  of  1839,  the  Centenary  of  which  we  are  com- 
memorating this  year,  has  developed  into  the  ubiquitous  photography 
of  today,  invading  every  field  of  art,  science,  and  industry,  of  which 
the  motion  picture  is  such  a  brilliant  example — it  has  been  a  long  road 
to  Hollywood! 


THE  SURFACE  OF  THE  NEAREST  STAR* 
R.  R.  McMATH** 

Summary. — Taking  motion  pictures  of  celestial  phenomena  that  show  change 
is  not  as  simple  as  it  would  first  appear.  Work  on  spectroheliokinematography  was 
started  in  1928,  and  in  1931  the  instrumentation  was  donated  to  the  University 
of  Michigan  by  the  founders  of  the  McMath-Hulbert  Observatory. 

The  combined  tower  telescope  and  spectroheliokinemato graph  of  this  observatory, 
now  one  of  the  most  powerful  pieces  of  solar  apparatus  in  the  world,  is  described  and 
some  of  the  solar  data  obtained  with  it  are  discussed. 

The  atoms  in  the  surfaces  of  all  the  billions  of  stars  accessible  with 
our  telescopes  may  fittingly  be  compared  to  minute  sending  stations, 
broadcasting  each  on  its  appointed  multitude  of  narrow  wave-length 
bands,  preserving  their  allotted  "channels"  with  almost  infinite  ex- 
actitude and  endeavoring  thus  to  send  us  certain  important  mes- 
sages relating  to  the  temperature  and  the  constitution  of  the 
stars  in  which  they  are  located.  With  out  telescopes,  and  to  a  far 
greater  extent  with  those  optical  receiving  stations  we  call  spectro- 
graphs,  we  are  today  reading  a  few  of  the  messages  these  distant  stars 
are  endeavoring  to  broadcast  to  us. 

Certain  equally  important  messages  relating  to  the  actual  spatial 
behavior  and  physical  movements  in  stellar  surfaces  seem,  however, 
permanently  beyond  our  reach.  For  no  telescope,  existing,  projected, 
or  imaginable,  can  show  a  star  to  us  as  anything  but  a  diskless  point 
of  light;  an  actual  stellar  diameter  of  a  million  miles  or  more  will 
vanish  almost  into  a  mathematical  point  at  stellar  distances  of  tril- 
lions or  quadrillions  of  miles. 

Thus  it  becomes  very  fortunate  for  our  knowledge  of  the  distant 
stars  that  we  have  a  star  so  close  at  hand  that  we  can  see  its  disk 
and  study  the  actual  motions  on  its  surface.  The  star  referred  to  is, 
of  course,  our  own  sun,  a  mere  bagatelle  of  ninety-three  millions  of 
miles  distant,  and,  fortunately  for  the  truth  of  the  deductions  we  may 
make  from  it  as  to  the  surface  behavior  of  stars  in  general,  a  respect- 

*  Presented  at  the  1938  Fall  Meeting  of  Detroit,  Mich. ;    reprinted  from 
Scientific  Monthly  (Nov.,  1938)  p.  411. 

**  The  McMath-Hulbert  Observatory,  University  of  Michigan. 
264 


SURFACE  OF  THE  NEAREST  STAR  265 

able,  run-of-the-mill,  middle-aged  star,  neither  very  hot  nor  very  cool 
as  stars  go,  and  neither  a  giant  nor  quite  a  dwarf  among  its  millions 
of  brother  suns.  The  entirely  average  position  of  our  sun  as  to  size, 
luminosity,  mass,  and  other  characteristics  thus  facilitates  and  makes 
more  probable  any  deductions  we  may  wish  to  draw  from  it  in  ap- 
plication to  more  distant  suns.  To  repeat,  any  study  of  the  stars  of 
our  universe  must  start  with  and  be  based  upon  a  study  of  our  nearest 
star — the  sun. 

About  twelve  years  ago  the  writer,  with  two  most  helpful  col- 
leagues— Judge  Henry  S.  Hulbert  and  the  late  Francis  C.  McMath, 
my  father — decided  that  a  fallow  and  hitherto  neglected  field  lay 
invitingly  open  for  research  through  the  application  of  the  motion 
picture  to  such  astronomical  phenomena  as  exhibit  rapid  motion  or 
change.  A  small,  but  most  completely  equipped  telescope  was  de- 
signed and  built  for  the  highly  exacting  technic  of  the  motion  picture 
as  applied  to  astronomical  photography,  and  this  installation  was 
located  at  Lake  Angelus,  about  five  miles  to  the  north  of  Pontiac, 
Mich.  The  instrumental  equipment  was  gradually  augmented  and 
improved  through  several  years  of  gradual  evolution  and  develop- 
ment, too  long  and  too  technical  to  detail  here,  and  in  1931  the  plant, 
under  the  name  of  the  McMath-Hulbert  Observatory,  was  deeded 
by  its  founders  to  the  University  of  Michigan. 

Our  initial  aims  were  frankly  educational.  We  envisaged  the  mani- 
fold assistance  that  carefully  planned  astronomical  films  would  give 
to  the  work  of  astronomical  instruction  in  schools  and  colleges.  How 
much  more  effective  it  would  be,  we  reasoned,  to  project  for  a  class  a 
three-minute  film,  showing,  for  example,  the  rotation  of  the  planet 
Jupiter  on  its  axis  and  the  revolutions  of  its  moons  about  the  planet, 
than  merely  to  lecture  to  a  class  that  such  things  were  happening. 
Many  thousand  feet  of  such  educational  films  were  taken  by  the  Mc- 
Math-Hulbert Observatory  of  planets  and  their  satellites,  the  phe- 
nomena of  sunrise  and  sunset  on  the  slowly  rotating  moon,  and  simi- 
lar subjects,  and  a  considerable  number  of  educational  reels  of  such 
types  have  been  shown  to  scientific  societies  and  distributed  to  schools 
and  colleges. 

It  is  with  a  feeling  of  some  regret  that  we  have  had  to  drop  most 
of  our  efforts  to  provide  purely  instructional  adjuvants  for  astronomi- 
cal teaching — we  hope  only  temporarily — for  we  are  still  firmly  con- 
vinced as  to  the  great  value  of  such  astronomical  films  for  the  in- 
structor as  well  as  for  the  student. 


266 


R.  R.  McMATH 


[J.  S.  M.  P.  E. 


The  reason  for  this  temporary  abandonment  is  comparatively 
simple;  it  has  come  about  merely  because  a  further  extension  of  this 
motion  picture  technic  to  that  nearest  star  we  call  the  sun  has  opened 
up  such  new  and  astonishingly  inviting  fields  of  scientific  research 
that  we  have  been  compelled,  willy-nilly,  to  devote  every  waking 
moment  to  a  new  and  fascinating  field  of  most  useful  scientific 
work  on  the  sun — the  actual  depiction  of  the  storms  around  sun-spots, 
and  the  intricacy  of  the  motions  of  the  mighty  gaseous  prominences 


FIG. 1. 


The  McMath-Hulbert  Observatory  of  The  University  of  Michigan, 
from  a  photograph  by  Sidney  D.  Waldon. 


that  rise  for  many  thousands  of  miles  above  the  solar  surface,  and 
move  and  change  and  disintegrate  with  speeds  that  range  from  a  few 
miles  per  second  up  to  explosive  velocities  of  several  hundred  miles 
per  second.  The  many  puzzles  which  are  exhibited  by  these  new 
pictures,  some  of  which  remain  as  yet  unsolved,  force  us  to  the  conclu- 
sion that  our  initial  purely  educational  aim  must  give  ground  for 
the  present  to  a  program  of  pure  scientific  research. 

There  are  several  respects  in  which  the  new  motion  pictures  of 
solar  phenomena  are  unique,  and  we  may  be  pardoned  for  assembling 
certain  of  their  outstanding  characteristics  at  this  point. 


Mar.,  1939]  SURFACE  OF  THE  NEAREST  STAR  267 

(1)  These  pictures  are  in  a  very  real  sense  "modern,"  inasmuch  as 
the  first  solar  films  taken  with  the  new  McMath-Hulbert  tower  tele- 
scope were  made  on  July  2,  1936,  one  day  after  the  completion  of  the 
tower. 

(2)  They  are  definitely  unique,  because  no  other  installation  at 
present  exists  which  has  the  instrumentation  for  similar  motion  pic- 
ture records  of  solar  phenomena. 

(3)  They  were  the  most  nearly  continuous  records  of  solar  phe- 
nomena ever  made,  and  in  this  factor,  as  will  be  noted  below,  lies 
perhaps  the  largest  portion  of  their  value  as  scientific  documents  for 
research  purposes.    Photographs  of  the  solar  surface  in  white  light, 
and  spectroheliograms  of  solar  prominences  in  the  light  of  calcium 
or  hydrogen  have  been  taken  for  several  decades,  but  most  of  these 
were  effectively  "stills,"  to  use  the  terminology  of  the  motion  picture 
studio.    Such  stills,  taken  at  time-intervals  of  an  hour  or  less,  have 
given  valuable  data  as  to  the  changes  occurring  in  solar  features; 
the  continuous  character  of  these  new  records  shows,  however,  the 
changes  as  they  are  taking  place,  and  not  only  make  possible  a  more 
detailed  study  of  the  mechanisms  underlying  the  phenomena,  but 
also  have  brought  to  light  a  mass  of  new  details,  hitherto  unsuspected, 
and  unrecorded  in  the  still  pictures  of  the  past. 

There  are  in  the  world  to-day  seven  tower  telescopes  for  studies 
of  the  sun;  that  at  Lake  Angelus  is  not  only  the  most  recent,  but 
embodies  many  refinements  of  design.  This  instrument  may  be  suc- 
cinctly described  as  a  telescope  which  remains  fixed  in  a  vertical 
position,  with  an  arrangement  of  motor-driven  mirrors  at  the  top  of 
the  tower,  termed  a  coelostat,  to  follow  the  sun  as  it  moves  across  the 
sky  and  to  throw  its  image  vertically  downward;  the  Lake  Angelus 
instrument  is  approximately  fifty  feet  in  height.  The  various  mirrors 
in  this  optical  train  are  of  pyrex,  which  is  peculiarly  fitted  for  solar 
instruments  because  of  its  very  low  coefficient  of  thermal  expansion. 
These  mirrors  are  covered  with  a  thin  coating  of  aluminum,  deposited 
by  evaporation  in  a  vacuum;  due  to  this  use  of  mirrors  rather  than 
lenses,  we  have  an  achromatic  telescope  that  is  exceedingly  rapid 
photographically.  For  this  reason,  the  exposures  with  the  Lake 
Angelus  apparatus  may  be  made  very  short;  exposures  on  solar 
prominences  in  current  work  range  from  ten  to  thirty  seconds,  where 
most  other  installations  must  count  their  exposure  times  in  minutes 
rather  in  seconds ;  such  short  exposures  make  for  a  record  that  is  prac- 
tically continuous. 


268  R.  R.  MCMATH  [j.  S.  M.  P.  E. 

At  the  bottom  of  the  tower  the  solar  image  formed  by  the  mirror 
train  falls  upon  the  slit  of  a  spectroheliograph.  This  rather  technical 
instrument  may  be  briefly  described  for  the  layman  as  a  spectrograph 
which  passes  the  light  from  the  solar  image  through  a  narrow  slit  and 
then  through  a  lens  to  a  grating  or  prism  which  is  located  in  a  heavy 
rotatable  steel  cage  in  a  well  thirty-five  feet  deep  beneath  the  tower. 
The  grating  disperses  and  spreads  out  the  light  in  the  form  of  a  spec- 
trum and  reflects  this  spectrum  through  the  same  lens  back  to  the 
upper  end.  Here  a  second  slit  is  installed  that  is  the  "heart"  of 
the  apparatus.  With  this  second  slit  we  pick  out  some  one  particular 
wavelength  of  some  element  in  the  solar  spectrum  and  throw  all  the 
rest  of  the  solar  light  away.  Solar  prominences,  for  example,  are  par- 
ticularly rich  in  the  elements  calcium  and  hydrogen.  Thus  we  may 
pick  out  one  definite  wavelength  of  calcium  and  secure  a  photograph 
of  a  narrow  strip  of  the  sun  where  the  solar  image  falls  upon  the  narrow 
slit,  in  calcium  light  only,  where  an  ordinary  photograph  in  "white" 
light  would  show  nothing,  because  of  the  overpowering  brilliance  of  the 
light  coming  from  other  chemical  elements  in  the  sun.  But  a  photo- 
graph of  a  narrow  strip  of  the  sun  in  calcium  light  would  be  useless 
and  almost  meaningless;  what  we  need  is  a  calcium  or  a  hydrogen 
light  picture  of  a  considerable  area,  either  of  the  solar  disk  itself  or 
of  an  area  of  the  solar  limb  where  some  large  prominence  is  seen  in 
profile.  To  secure  such  a  picture  of  an  area,  the  first  slit  is  moved 
back  and  forth  over  the  chosen  area  of  the  solar  image  and  the  second, 
or  "picking-out"  slit  is  given  a  precisely  equal  but  exactly  opposite 
motion  so  as  always  to  receive  the  calcium  wavelength  of  the  spec- 
trum reflected  from  the  grating,  and  that  wavelength  only. 

The  result  of  this  scanning  process,  performed  twice  a  second,  is  a 
calcium  or  a  hydrogen  picture  of  an  area.  Some  other  elements  may 
be  selected  as  well,  in  case  we  wish  to  secure  an  iron  picture  or  a 
helium  picture,  and  all  these  pictures  in  the  light  of  some  chosen  ele- 
ment would  ordinarily  be  entirely  invisible,  but  are  made  possible 
only  by  this  process  of  sorting  out  a  definite  wavelength  and  discard- 
ing all  the  rest  of  the  light  from  the  sun. 

The  above  brief  and  schematic  description,  manifestly,  can  give 
but  a  slight  idea  of  the  actual  complexity  of  the  apparatus,  some  con- 
ception of  which  may  be  derived  from  the  fact  that  there  are  about 
forty  small  electric  motors  scattered  over  the  tower  mechanisms  from 
the  coelostat  at  the  top  to  the  grating  cage  down  in  the  well,  and  each 
of  these  is  controlled  by  its  individual  push-button.  In  these  re- 


Mar.,  1939]  SURFACE  OF  THE  NEAREST  STAR  269 

spects  the  present  installation  is  doubtless  the  most  convenient  in  ex- 
istence, as  the  observer  at  the  spectroheliograph  head  can  perform  any 
adjustment  or  manipulation  without  leaving  his  station,  merely  by 
pressing  an  electric  push-button. 

The  apparent  complexity  of  certain  features  of  the  mechanism  of 
this  tower  telescope  is,  in  some  senses,  merely  a  necessary  consequence 
of  the  exacting  technic  that  has  been  found  indispensable  for  the  tak- 
ing of  satisfactory  motion  pictures  of  this  and  other  celestial  phe- 
nomena. A  "run"  or  "scene"  may  comprise  anything  from  a  few 
hundred  to  over  a  thousand  separate  pictures  on  the  film;  the  word 
"frame"  is  customarily  used  for  these  individual  pictures;  six  or  eight 
hours  of  continuous  work  will  ordinarily  go  into  a  run  comprising  a 
thousand  separate  frames. 

Manifestly,  all  the  frames  of  a  scene  must  be  as  perfectly  registered 
as  possible,  to  avoid  flickering  and  unsteadiness  on  the  screen.  Early 
in  the  work  on  the  moon  and  planets  that  preceded  this  solar  work,  it 
was  found  that  no  existing  form  of  telescope  drive  gave  sufficient 
accuracy.  Accordingly,  merely  as  a  by-product  of  the  larger  pro- 
gram, and  after  four  other  methods  had  been  tried  and  found  want- 
ing, a  new  and  improved  form  of  telescope  drive  was  devised,  based 
upon  an  infinitely  flexible  and  instantly  variable  control  of  the  input 
electrical  frequency  to  the  telescope  drive  motor,  secured  through 
resistance-ballasted  thermionic  tubes.  This  form  of  telescope  drive 
is  known  as  the  McMath-Hulbert  electric  drive ;  it  brings  it  to  pass 
that  the  telescope  becomes  an  automatically  following  instead  of  a 
manually  guided  apparatus;  it  has  since  been  adopted  for  the  drive 
of  the  McDonald  reflector  in  Texas,  for  three  telescopes  at  Lick  Ob- 
servatory, and  is  under  consideration  for  other  projected  large  telescope 
mountings.  The  instruments  at  Lake  Angelus  were  also  the  first  to 
employ  a  similar  accurately  controlled  drive  in  the  declination  com- 
ponent, in  addition  to  the  ordinary  motion  given  in  the  right  ascension 
coordinate. 

In  the  astronomical  motion  picture  technic,  it  must  also  be  possible 
to  arrange  for  any  probable  desired  duration  of  the  actual  exposure, 
as  well  as  for  the  duration  of  the  "dark  time"  between  exposures.  A 
gearing  train  in  an  underground  control  room  adjacent  to  the  tower 
makes  possible  the  selection  of  these  times  and  controls  the  shutter 
of  the  special  motion  picture  camera.  A  description,  or  even  a  bare 
tabulation,  of  all  the  necessary  mechanical  details  is,  however,  mani- 
festly impossible  in  a  general  article.  Only  one  additional  desidera- 


270  R.  R.  McMATH  [J.  S.  M.  P.  E. 

turn  may  be  noted.  The  hundreds  of  separate  frames  in  a  scene 
would  have  scant  scientific  value  as  records  of  motion  and  change  if 
accurate  timing  arrangements  were  not  provided.  Accordingly  each 
individual  frame  automatically  makes  a  record  of  its  time  electrically 
on  a  continuously  running  chronograph  in  the  underground  control 
room. 

With  an  exposure  of  twenty-seven  seconds  on  a  solar  prominence 
and  a  dark  time  of  three  seconds,  two  frames  will  be  taken  per  minute ; 
they  will  be  projected  on  the  screen  at  the  customary  rate  of  sixteen 
per  second,  or  960  frames  per  minute.  Thus  it  will  be  evident  that  the 
projected  picture  will  have  a  "compression  factor"  of  1 :480.  Such  a 
compression  of  the  record  is  not  only  inevitable,  but  a  very  distinct 
advantage,  rather  than  a  detriment. 

Suppose,  as  is  not  at  all  unusual,  that  a  bright  knot  is  seen  to  form 
100,000  miles  above  the  solar  surface  and  then  to  descend  at  the 
rate  of  40  miles  per  second,  about  average  as  solar  prominence  ve- 
locities run,  but  still  80  times  the  speed  of  a  high-power  rifle  bullet. 
Its  total  time  of  descent  to  the  sun  will  be  forty-two  minutes.  In- 
stead of  having  to  wait  that  long  in  our  seats  to  see  the  history  of  this 
descending  knot,  the  above  compression  factor  of  1 : 480  reduces  it  to 
about  5  seconds,  and  a  scene  which  is  made  up  of  several  such  knots 
in  motion  will  occupy  the  very  convenient  interval  of  20  or  25 
seconds  and  appears  practically  continuous  in  its  record  of  motions 
and  changes. 

By  design,  considerable  space  has  here  been  given  to  an  outline  of 
the  technic  of  the  motion  picture  as  adapted  to  an  astronomical  end, 
and  the  apparatus  necessary  for  the  purpose,  in  order  to  emphasize, 
not  only  the  more  difficult  features  of  the  research  on  its  mechanical 
side  but  also  the  unique  character  of  the  resulting  record. 

During  the  seasons  of  1936  and  1937  over  ten  thousand  feet  of 
standard  35-mm.  film  were  exposed  in  the  new  tower  telescope  on 
the  solar  prominences  or  on  features  of  the  solar  disk  itself.  Even 
though  every  possible  mechanical  convenience  or  electrical  adjust- 
ment has  been  provided,  and  even  though  this  tower  telescope  has 
been  pronounced  to  be  the  most  rapid,  flexible,  and  convenient  in 
existence,  the  total  amount  of  labor  and  attention  involved  in  taking 
over  one  thousand  separate  photographs  in  the  run  on  a  clear  day 
which  may  extend  from  8  A.M.  or  earlier  till  6  P.M.  is  very  considerable. 
It  is  a  pleasure  to  make  acknowledgment  at  this  point  to  my  two  col- 
leagues and  to  those  who  have  assisted  in  the  somewhat  complicated 


Mar.,  1939] 


SURFACE  OF  THE  NEAREST  STAR 


271 


FIG.  2.     The  50'  tower  telescope  of  the  McMath-Hulbert  Observatory  from 
a  drawing  by  Russell  W.  Porter. 

technic  of  solar  prominence  photography  and  measurement — to  our 
research  associate,  Dr.  Edison  Pettit,  of  the  Mount  Wilson  Observa- 
tory, and  to  Harold  E.  Sawyer,  assistant  astronomer,  and  John 
Brodie,  assistant,  in  the  McMath-Hulbert  Observatory;  others  have 
given  assistance  for  shorter  periods.  We  owe  also  a  special  debt  of 


272  R.  R.  McMATH  [J.  S.  M.  P.  E. 

thanks  to  Dr.  Heber  D.  Curtis,  director  of  the  observatories  of  the 
University  of  Michigan,  who  has,  from  its  very  inceptance,  given 
every  encouragement  to  this  program  of  solar  research  and  every 
assistance  within  his  power. 

These  films,  when  projected  under  proper  conditions,  show  scenes 
of  unexampled  grandeur,  and  radically  change  our  preconceived  no- 
tions of  the  surface  of  a  star.  Though  we  knew  from  the  "still" 
photographs  of  the  past  that  the  sun's  surface  was  marked  by  con- 
stant activity  as  manifested  by  those  solar  storms  called  sun-spots,  by 
the  flocculi  and  by  the  prominences,  these  films  for  the  first  time 
bring  to  us  the  actual  motions  in  a  continuous  record,  which  we  may 
repeat  as  often  as  we  need  for  our  scientific  studies.  These  motion 
pictures  very  effectively  change  our  conception  of  a  star's  surface 
from  something  at  least  relatively  static  to  a  picture  that  is  intensely 
kinetic;  we  begin  to  realize  that  the  surface  of  a  star  is  an  unending 
maelstrom  of  motions  due  to  titanic  forces  whose  precise  nature  can 
not  as  yet  be  regarded  as  completely  explained. 

Even  though  we  are  astronomers,  we  are  very  human,  and  we  too 
derive  much  the  same  pleasure  as  does  the  layman  who  sees  these 
films  and  is  enthusiastic  in  his  praise  of  them,  viewed  merely  as  in- 
spiring spectacles.  And  yet,  strange  as  it  may  seem,  we  who  are  tak- 
ing and  studying  these  new  records  of  solar  activity  take  rather  amiss 
the  enthusiastic  praises  we  hear  from  laymen  or  from  scientists  in 
other  fields  who  apparently  regard  them  as  merely  interesting 
"movies."  We  feel  quite  strongly  that  the  magnificence  of  these 
displays  is,  in  many  respects,  only  a  very  secondary  consideration  in 
our  evaluation  of  these  pictures  as  scientific  records,  from  which  facts 
of  very  definite  value  are  being  derived  as  to  the  actual  nature  of 
the  surface  of  a  star. 

"Conflagrations,"  explosions,  skyrocket  sheaves  of  light  like  the 
grand  finale  of  the  4th  of  July  celebrations  of  our  boyhood,  are  all 
admittedly  inspiring  when  we  realize  the  tremendous  speeds  that  are 
actually  involved,  the  temperature  of  more  than  10,000°F,  that  our 
pictures  embraces  an  area  150,000  miles  high  and  200,000  miles  wide, 
and  that  our  earth  would  be  but  a  small  disk  on  the  same  scale  and 
quite  unimportant  in  comparison  to  the  mighty  flames  and  streamers 
of  incadescent  gas  that  form  these  solar  storms.  Yet  to  us  the  mo- 
tions and  laws  of  motion  that  we  are  deriving  and  the  nature  of  the 
mysterious  forces  that  seem  eternally  operative  on  the  surface  of  a 


Mar.,  1939]  SURFACE  OF  THE  NEAREST  STAR  273 

star  are  much  the  more  important  considerations  as  we  view  these  new 
films. 

This  new  method  of  attack  on  the  problems  exhibited  by  the  surface 
of  a  star  is  still  too  youthful  to  admit  of  explanations  of  each  and 
every  phenomenon  observed.  Fresh  puzzles  too  frequently  show 
themselves  in  each  run  on  a  new  and  active  prominence,  and  if  the 
history  of  our  past  work  is  any  criterion,  the  coming  season  of  1938 
will  bring  to  light  as  many  new  features  as  have  those  of  the  two 
preceding  years.  The  complexity  of  some  of  the  more  active  promi- 
nence displays  frequently  baffles  description,  and  we  often  find  that 
the  only  way  to  be  sure  of  all  that  is  taking  place  is  a  repeated  showing 
of  the  film ;  frequently  we  will  notice  some  minor  peculiarity  or  puzzle 
in  the  tenth  or  twelfth  showing  that  had  previously  escaped  us  and 
had,  of  course,  never  been  suspected  in  the  still  pictures  of  the  past. 

While,  as  already  noted,  we  are  still  working  on  many  of  the  puzzles 
presented,  and  are  withholding  a  more  precise  formation  of  hypotheses 
till  more  data  have  been  collected,  some  of  the  results  of  the  work  on 
the  sun  with  the  new  tower  telescope  and  our  improved  motion  pic- 
ture technic  may  be  assembled  as  follows,  either  in  more  general 
statements  or  in  descriptions  of  isolated  phenomena. 

(1)  It  has  become  necessary  to  add  three  subdivisions  to  Dr. 
Pettit's  accepted  classification  scheme  for  solar  prominences,  to 
include  three  new  types  of  prominence  whose  existence  was  not  hither- 
to suspected.  These  are: 

(a)  Surges. — These  are  very  short-lived  prominences  like  spear- 
heads of  flame,  that  stab  upward  1000  to  10,000  miles  or  so  from  the 
solar  chromosphere  and  as  rapidly  subside  again,  with  a  total  life 
period  of  only  a  few  minutes.    As  seen  in  profile  in  runs  on  promi- 
nences, the  limb  of  the  sun  will  occasionally  exhibit  an  almost  con- 
tinuous activity  of  this  type.    In  pictures  of  the  solar  disk  proper, 
the  sudden  short-lived  splotches  of  brilliant  light  that  appear  and  dis- 
appear in  areas  about  sun-spots  are  believed  to  be  these  same  surges, 
seen  from  above. 

(b)  Ejections. — This  name  has  been  given  to  the  balls  of  luminous 
chromospheric  matter  thrown  out  of  sun-spot  areas  like  Roman- 
candle  displays.     They  are  relatively  faint  and  seem  to  leave  the 
sun  without  returning.    In  one  or  two  cases  these  balls  seem  rather 
more  like  hollow  spheres  or  perhaps  in  the  form  of  smoke  rings;  it  is 
difficult  to  decide  with  the  material  now  available. 

(c)  Coronal  Type  Streamers. — These  are  very  puzzling.     In  such 


274 


R.  R.  McMATH 


LT.  S.  M.  P.  E. 


FIG.  3.       Great  eruptive  prominence  of  September  17,   1937,  photo- 
graphed at  The  McMath-Hulbert  Observatory. 

A.     14h50.m69  B.     14h55.m84  C.     15h06.m13     GCT 

D.    15h09.mll  E.     15h14.m31  F.     16h06.m  7    GCT 

Exposures  A  to  E  with  20-ft.  focus  mirror,  F  with  lens  of  74  inches  focal 
length.  In  "F"  the  prominence  goes  out  of  the  picture  1,000,000  km. 
above  the  sun. 


streamers  matter  appears  to  form,  or  more  properly  to  become  lumi- 
nous, at  an  altitude  of  120,000  miles  or  more  above  the  surface  of  the 
sun  and  then  to  descend  in  successive  streamers  to  the  solar  surface. 
These  coronal  streamers  are  generally  rather  faint  and  have  never 
been  detected  before. 


Mar.,  1939]  SURFACE  OF  THE  NEAREST  STAR  275 

(2)  Several  very  interesting  examples  of  violently  eruptive  promi- 
nences have  been  recorded.     For  one  of  these,  taken  in  September, 
1937,  though  the  entire  period  covered  by  the  scene  was  only  80 
minutes,  the  upper  portions  could  not  be  kept  within  the  motion 
picture  frame  in  spite  of  three  successive  changes  to  shorter  focal 
lengths;   the  total  height  was  about  620,000  miles,  which  held  the 
world's  record  until  the  recent  record  of  900,000  miles  for  a  promi- 
nence photographed  at  Mount  Wilson.     The  velocity  of  this  Lake 
Angelus  prominence  reached  432  miles  per  second ;  as  this  is  consider- 
ably greater  than  the  "velocity  of  escape"  under  gravitational  attrac- 
tion at  this  distance  from  the  sun,  this  is  believed  to  be  the  first 
recorded  instance  where  we  have  observed  matter  shot  out  into  space 
beyond  the  sun's  attraction,  though  such  possibilities  have  long  been 
recognized  in  theory. 

(3)  Arch  Types. — Several  great  arches  of  unusual  interest  have  been 
recorded.    In  one  of  these  it  was  nearly  100,000  miles  between  the 
"feet  of  the  rainbow."    Though  there  was  no  noticeable  accretion  of 
material  at  the  top  of  this  arch,  luminous  knots  of  gas  are  observed 
continuously  descending  to  the  sun  in  both  directions  from  the  summit 
on  the  arch.    Why? 

(4)  Predominance  of  Matter  in  Descent. — Even  if  we  include  the 
prominences  of  eruptive  type  mentioned  under  2  above,  perhaps  90 
per  cent  of  our  prominence  scenes  record  matter  in  descent  only.    On 
a  number  of  great  "banyan-tree"  prominences,  with  multiple  stalks  or 
trunks  connecting  the  enlarged  upper  portions  to  the  chromosphere, 
bright  nodules  of  matter  will  be  observed  spiralling  downward  along 
the  "trunks."    We  have  mentioned  above  under  Ic  the  growth  lumi- 
nescence in,  or  actual  formation  of  faint  clouds  high  above  the  sun, 
from  which  the  coronal  type  streamers  descend,  phenomena  which 
seem  to  necessitate  the  postulation  of  some  form  of  solar  chromo- 
spheric  atmosphere  intermixed  with  the  corona. 

Much  the  same  class  of  phenomena  are  exhibited  in  lower  bright 
streamers  of  the  beautiful  "set-pieces"  of  a  fountain  type;  the  mo- 
tion of  descent  is  here  often  clear  and  rapid ;  any  corresponding  as- 
cent of  matter  on  a  possible  rising  arm  of  the  complete  trajectory  is 
either  very  much  fainter  or  entirely  absent.  Astronomers  who  see 
these  films  for  the  first  time  frequently  attempt  to  explain  this  curious 
phenomenon  by  ascribing  the  invisibility  of  the  ascending  side  of  the 
streamer  to  the  Doppler  effect,  arguing  that  some  velocity  in  the  line 
of  sight  moves  the  wavelength  under  observation  "off  the  slit"  for  the 


276  R.  R.  McMATH  [J.  S.  M.  P.  E. 

ascending  branch  and  implying  that  these  films  give  a  partial  rather 
than  a  true  picture  of  these  motions.  We  are  utterly  unable  to  accept 
this  explanation  in  the  vast  majority  of  cases,  for  we  have  noted  only 
two  cases  of  sudden  brightening  of  small  patches  near  the  chromo- 
sphere that  may  possibly  be  due  to  the  Doppler  effect,  that  is,  to  a 
velocity  in  the  line  of  sight  sufficient  to  bring  a  different  wavelength 
and  hence  a  previously  unobserved  detail  into  the  slit  of  the  instru- 
ment. But  an  elementary  consideration  of  the  geometry  of  these 
prominence  arches  would  predicate  roughly  equal  velocities  in  the 
line  of  sight  for  matter  at  corresponding  portions  of  the  hypothetical 
ascending  or  the  descending  branches  of  the  arch.  Moreover,  al- 
though workers  in  the  past  have  maintained  that  narrow  slits  are 
an  inescapable  necessity  in  spectroheliographic  work,  we  have,  as  a 
result  of  numerous  experiments  with  wider  slits,  taken  beautifully 
clear  and  sharp  spectroheliograms  in  the  H  alpha  line  of  hydrogen  with 
both  slits  0.5  mm.  (about  one  fiftieth  of  an  inch)  in  width.  This  width 
would  necessitate  a  difference  in  radial  velocity  of  about  ±110  km.  per 
second  (68  miles  per  second)  to  move  a  portion  of  our  picture  off  the 
slit  and  thus  render  some  parts  invisible.  While  higher  velocities  are 
occasionally  observed  in  eruptive  prominences,  such  velocities  have 
only  rarely  been  found  in  these  arch  trajectories. 

The  phenomenon  remains  a  puzzle.  It  is  apparent  that  what  goes 
down  very  probably  came  up,  but  why  should  the  upward  journey  be 
predominantly  invisible?  Is  it  that  the  gases  on  their  upward  path 
are  in  some  different  temperature  or  ionization  state,  changing  back 
to  another  and  photographically  recordable  state  soon  after  passing 
the  crest  of  their  trajectory?  Data  are  being  collected  that  may  even- 
tually give  an  answer  to  this  pressing  and  difficult  question. 

(5)  Previous  work  had  detected  no  motions  within  the  dark  hy- 
drogen flocculi.    We  have  been  fortunate  enough  to  "catch"  and  to 
photograph  for  a  total  elapsed  time  of  5x/2  days  an  enormous  hydrogen 
flocculus  whose  total  length  must  have  been  of  the  order  of  700,000 
miles,  extending  over  a  considerable  portion  of  the  solar  disk.     Its 
internal  motions  and  its  final  disintegration  were  clearly  recorded. 
So  far  as  is  known,  this  is  the  first  case  where  the  life-history  of  one 
of  these  dark  hydrogen  flocculi  has  been  followed  from  its  first  appear- 
ance to  the  end. 

(6)  Abundant  confirmation  has  been  secured  in  the  study  of  mo- 
tions in  prominence  streamers  in  support  of  the  curious  laws  of 
prominence  motion  discovered  earlier  by  Dr.  Pettit.    The  velocity  of  a 


Mar.,  1939]  SURFACE  OF  THE  NEAREST  STAR  277 

prominence  or  prominence  formation  is  uniform,  increasing  suddenly 
at  intervals.  When  there  is  a  change  in  velocity  the  new  velocity  is 
generally  a  simple  multiple  of  the  previous  velocity.  That  is,  a  knot 
that  has  been  moving  along  a  streamer,  at  a  uniform  speed  of,  say,  21 
miles  per  second  will  suddenly  (sometimes  in  less  than  a  minute)  be 
accelerated  to  a  uniform  speed  of  42  miles  per  second,  without  any 
apparent  transition  through  intermediate  velocities.  This  puzzling 
phenomenon  has  been  studied  and  extended  to  include  nearly  all 
prominence  types  at  Lake  Angelus.  Like  some  others  found  in  the 
Lake  Angelus  work  mentioned  above,  it  shows  that  we  have  many 
still  unknown  factors  with  which  to  deal  before  we  can  secure  an  ex- 
planation of  all  the  laws  that  govern  the  motion  of  gaseous  matter 
near  the  surface  of  a  star. 

Such,  then,  are  some  of  the  results,  as  well  as  some  of  the  problems, 
that  are  growing  out  of  the  application  of  this  new  technic  to  the 
study  of  a  star's  surface  behavior;  the  work  is  being  continued  as 
fast  as  time  and  money  will  permit.  The  very  richness  of  the  material 
being  secured  has  its  embarrasing  features;  it  will  easily  be  seen  that 
the  detailed  measurement  of  the  motions  even  in  one  tenth  of  the 
frames  of  a  prominence  picture  that  includes  over  a  thousand  separate 
pictures  involves  a  great  deal  of  time  and  not  a  little  calculation.  The 
number  of  the  prominences,  as  well  as  the  number  and  the  activity 
of  the  flocculi  and  other  disk  features,  shows  an  intimate  connection 
with  the  curve  of  the  number  of  sun-spots  that  reaches  a  maximum 
roughly  every  eleven  and  a  third  years.  We  have  recently  been 
passing  through  a  period  of  maximum  sun-spot  activity,  but  we  do 
not  yet  know  what  detail  changes  in  prominence  activity  will  be  ob- 
served on  our  films  at  a  sun-spot  minimum.  We  are  inclined  to  pre- 
dict that  while  we  shall  then  have  fewer  prominences  on  which  to 
work,  they  still  will  be  of  equal  value  in  formulating  theories  of  the 
surface  layers  of  a  star.  Certainly  only  a  beginning  of  our  program  of 
research  will  have  been  made  until  we  have  worked  completely 
through  at  least  one  sun-spot  cycle. 

At  the  close  of  his  presentation,  Mr.  Me  Math  showed  motion  pictures  of  the  sun 
taken  with  the  spectroheliokinematograph  described  in  the  paper.  The  motion 
pictures  showed  the  prominences  and  other  phenomena  of  the  sun  taken  under  hy- 
drogen and  calcium  light.  The  types  of  prominences  and  streamers  are  described 
in  the  paper.  Sun  spots  and  other  eruptions  and  ejections  were  shown. 

DISCUSSION 

MR.  McMATH:  Please  bear  in  mind  that  you  were  looking  at  atoms,  and 
do  not  try  to  interpret  what  you  have  seen  in  terms  of  molecular  physics.  The 


278  R.  R.  McMATH  [J.  S.  M.  P.  E. 

temperatures  are  at  least  10,000  °F,  and  as  far  as  we  know  there  are  few  mole- 
cules or  atoms  in  association  at  that  temperature.  Of  course,  we  are  photo- 
graphing a  very  simple  atom. 

MR.  CRABTREE:    How  long  do  the  eruptions  last? 

MR.  McMATH:  The  duration  is  measured  in  either  days  or  minutes,  the 
shortest  being  about  20  minutes. 

MR.  KELLOGG:  Mr.  McMath  has  given  us  a  real  thrill.  How  nearly  the 
same  would  the  picture  be  if  it  were  taken  with  hydrogen  instead  of  calcium? 
Are  the  gases  localized?  Is  what  we  saw  a  phenomenon  in  hydrogen  gas  alone, 
or  calcium  alone? 

MR.  McMATH :  No,  it  appears  that  the  prominence  is  a  homogeneous  mixture 
of  gases.  They  actually  are  about  98  per  cent  hydrogen  gas,  and  the  rest  is 
principally  calcium  and  helium.  As  we  approach  the  chromosphere  we  commence 
to  get  traces  of  heavier  elements,  for  instance,  sodium.  We  have  simultaneous 
pictures,  many  thousands  of  feet  of  them,  taken  in  calcium  and  hydrogen,  at 
precisely  the  same  time,  and  the  two  pictures  look  practically  alike,  and  the  two 
gases  behave  alike  and  measure  alike. 

MR.  KELLOGG:  Is  what  appears  to  be  motion  really  regions  of  intense  pro- 
gressive ionization  of  gas  by  free  traveling  electrons  or  by  electromagnetic  waves, 
rather  than  the  motion  of  the  material  itself? 

MR.  McMATH:  It  is  a  very  long  story.  These  prominences  are  seen  at 
eclipses,  and  we  have  some  eclipse  photography  under  the  whole  integrated  light 
which  would  eliminate  some  spectroscopic  implications.  Since  we  have  done 
this  work  in  neutral  hydrogen  atoms,  as  well  as  the  ionized  calcium,  we  believe 
that  we  are  looking  at  the  streaming  of  individual  atoms. 

MR.  MATTHEWS:  How  thick  is  the  chromosphere?  Does  it  vary  in  thickness 
from  time  to  time? 

MR.  MCMATH:  It  will  average  from  between  5000  and  6000  kilometers  at 
eclipses.  When  a  picture  is  taken  with  white  light  with  an  ordinary  camera, 
we  get  a  picture  of  the  photosphere.  That  is  where  the  spots  show  up  black. 
Just  above  the  photosphere  is  a  layer  of  gases,  the  chromosphere.  This  is  the 
reversing  area,  from  which  we  get  the  dark  hydrogen  line  of  the  solar  spectrum, 
known  as  an  absorption  line.  In  the  prominences,  of  course,  we  take  the  picture 
in  the  emission  line  of  the  hydrogen. 

MR.  WOLF:  What  do  you  anticipate  can  be  done  with  the  new  200-inch  tele- 
scope? 

MR.  McMATH:  It  will  probably  never  be  used  on  the  sun.  I  am  on  the 
Advisory  Committee  for  the  200-inch  telescope,  and  I  wish  it  would  be  used  for 
the  sun,  but  it  will  not  be.  There  are  some  objections  to  solar  work  with  an 
equatorial  telescope,  not  easily  explained  in  a  few  minutes. 

MR.  JOHNSON:  In  the  course  of  the  day  is  there  a  sensible  change  in  the 
apparent  motion  of  the  sun? 

MR.  MCMATH:  That  is  due  to  refraction.  At  sunrise  and  sunset  we  get 
maximum  refraction.  At  sunrise  the  sun  appears  in  the  telescope  to  travel 
more  slowly  than  it  does  when  on  the  meridian  at  noon. 

MR.  MATTHEWS:  Have  you  any  explanation  of  why  the  prominences  are 
drawn  back  in? 

MR.  MCMATH:    No.    All  the  earlier  work  was  based  on  material  rising  from 


Mar.,  1939]  SURFACE  OF  THE  NEAREST  STAR  279 

the  sun.  An  English  group  tried  to  explain  the  prominences  by  radiation  pres- 
sure. Now  we  find,  in,  say,  one  hundred  hours  of  photography,  about  90  per  cent 
of  the  pictures  show  material  going  down  into  the  sun.  All  evidence  obtained 
at  eclipses  denies  specifically  the  existence  of  chromospheric  material  in  the 
corona  or  in  the  solar  atmosphere,  we  will  say.  However,  we  are  forced  into  the 
postulation  of  some  kind  of  solar  atmosphere.  The  material  is  certainly  there. 
It  would  not  appear  out  of  nothing. 

MR.  STROCK:  Are  these  prominences  the  disturbances  that  are  correlated 
with  radio  fading? 

MR.  McMATH:  Not  always.  That  problem  is  being  investigated  by  J.  H. 
Bellinger  of  the  Radio  Division  of  the  National  Bureau  of  Standards,  and  others. 
There  have  been  certain  chromospheric  eruptions  that  have  been  coincident  with 
general  short-wave  radio  fadeouts  on  the  sunlit  hemisphere.  We  do  not  know 
precisely  what  type  of  activity  is  associated  with  these  fadeouts.  Sometimes 
very  intense  activity  has  no  noticeable  effect  upon  radio  reception.  The  enor- 
mous eruption  of  September  17,  1939,  had  no  effect  upon  radio  reception  at  all, 
and  yet  it  was  very  intense. 

MR.  STROCK:  Do  we  know  whether  these  prominences  are  tied  in  with  the 
so-called  eleven-year  sun-spot  cycle? 

MR.  McMATH:  They  occur  with  solar  activity.  The  more  sun  spots,  the 
more  solar  activity;  and  the  more  solar  activity,  the  more  prominences. 


THE  ELECTRICAL  PRODUCTION  OF  MUSICAL  TONES5 

S.  T.  FISHER** 


Summary. — Music  is  a  study  that  is  primarily  an  art;  its  scientific  aspects  have 
been  recognized  since  ancient  days.  An  outline  is  given  of  the  physical  basis  of 
harmony  upon  which  in  turn  are  based  the  musical  scales.  A  short  outline  of  the 
historical  development  of  the  modern  scale  is  included,  the  discords  produced  in  the 
tempered  scale  due  to  natural  harmonics  and  summation  and  difference  tones  being 
noted.  In  the  electronic  instruments  the  musical  tones  can  be  generated  as  such  in  the 
first  instance,  or  may  be  produced  as  a  synthesis  of  pure  tones;  instruments  embody- 
ing these  principles  are  described  briefly.  A  method  of  the  synthesis  of  musical 
qualities  is  described,  and  at  the  presentation  of  the  paper,  such  tones  were  demon- 
strated using  the  Hammond  (electric)  organ,  the  wave-forms  being  shown  on  an 
oscillograph. 

In  conclusion  it  is  suggested  that  the  possibility  of  removing  the  limitations  of  the 
traditional  keyboard  instruments  gives  an  opportunity  of  abandoning  the  tempered 
scale  in  favor  of  the  just  scale. 

This  paper  deals  with  a  topic  that  lies  partly  in  the  realm  of  art 
and  partly  in  the  realm  of  science;  it  is  not  clear  where  one  ends  and 
the  other  begins.  If  the  question  is  debated  as  to  wherein  lies  the 
superiority  of  Beethoven  over  Puccini,  science  can  take  no  part.  The 
problem  is  either  too  vague  or  too  complicated  to  be  dealt  with  on  a 
precise  basis.  Physics  can  not  tell  us  why  we  find  Beethoven  par- 
ticularly pleasing  but  it  can  explain  why  an  error  made  in  trans- 
posing his  music  from  one  key  to  another  would  make  it  disagreeable. 

Let  us  first  look  at  the  elementary  facts  of  the  physical  basis  of 
music.  Music  is  an  art;  our  response  to  it  is  psychological  and  emo- 
tional. Nevertheless  if,  as  engineers,  we  set  out  to  produce,  to  trans- 
mit, and  to  reproduce  music,  we  are  forced  to  translate  the  intangible 
into  terms  of  the  tangible.  Just  as  a  knowledge  of  anatomy  is  basic 
in  the  training  of  a  sculptor  or  painter,  so  in  music  it  is  necessary  to 
understand  the  arithmetic  that  forms  the  groundwork  of  the  art. 

*Received  December  1,  1938;  presented  before  the  Montreal  Sections  of  the 
Engineering  Institute  of  Canada  and  the  Institute  of  Radio  Engineers,  Oct.  12, 
1938,  and  the  Toronto  Sections  of  the  Institute  of  Radio  Engineers  and  the  Ameri- 
can Institute  of  Electrical  Engineers,  Oct.  17,  1938. 
**Northern  Electric  Co.,  Ltd.,  Montreal,  Canada. 

280 


ELECTRICAL  PRODUCTION  OF  MUSICAL  TONES  281 

There  are  three  attributes  that  we  recognize  as  essential  in  music : 
rhythm,  melody,  harmony.  Rhythm,  the  maintenance  and  accenting 
of  the  time  intervals  at  which  sounds  are  produced,  is  the  simplest 
and  most  obvious  part  of  music,  and  is  the  chief  characteristic  of 
primitive  music.  Melody  is  the  linking  together  of  notes  of  different 
pitches  and  durations  in  a  sequence  that  is  pleasing,  harmonious, 
and  readily  capable  of  being  memorized.  It  is  the  attribute  of  music 
next  in  complexity  to  rhythm,  and  many  modern  types  of  music, 
e.  g.y  the  Persian,  consist  only  of  single  notes  played  in  sequence  and 
in  rhythm — the  kind  of  music  a  man  whistling  and  tapping  his  foot 
produces.  The  third  attribute,  harmony,  is  inextricably  bound  up 
with  melody  in  a  very  complicated  fashion,  and  is  the  art  and  science 
of  sounding  two  or  more  notes  simultaneously  so  that  a  pleasing 
effect  is  obtained,  of  greater  richness,  variety,  and  interest  than  a 
single  note  can  produce. 

It  is  plain  that  music  requires  a  fixed  series  of  notes  whose  pitches, 
*.  e.,  frequencies,  are  such  that  they  give  pleasing  effects  when  sounded 
in  various  sequences,  and  that  can  be  grouped  in  a  number  of  ways  and 
sounded  simultaneously  without  offending  the  ear.  What  then  are 
the  physical  realities  back  of  such  terms  as  "pleasing,"  "offending"? 
A  melody  sounds  well  when  its  component  notes  are  those  that  sound 
well  in  harmony.  This  leads  to  the  simple  rule  that  has  been  uni- 
versally recognized  for  at  least  30  centuries  as  the  fundamental  basis 
of  music:  Two  notes  sounded  simultaneously  are  harmonious  if  the 
ratio  of  their  frequencies  can  be  expressed  as  the  ratio  of  small  integral 
numbers. 

From  this  rule,  we  should  expect  that  notes  with  the  following 
frequency  ratios  would  be  harmonious,  the  degree  of  concord 
lessening  as  we  go  down  the  list : 

12345    3455 
-»-»-»-»  -» etc.,-*  -•  -»  -'  etc. 
11111    2334 

The  reason  why  the  combination  of  two  tones  is  harmonious  when 
the  tone  frequencies  have  a  simple  numerical  ratio  has  been  the  sub- 
ject for  conjecture  since  Pythagoras,  who  flourished  about  700  B.C. 
Pythagoras  considered  such  harmony  inherent  in  nature  and  theology; 
Confucius  also  solved  the  problem  by  a  relapse  to  metaphysics.  In 
the  18th  century  the  mathematician  Euler  brought  psychology  to 
bear,  and  attributed  the  concord  to  our  delight  in  method  and  order- 
liness, and  proposed  an  elaborate  system  long  since  discarded  for 


282  S.  T.  FISHER  tf.  S.  M.  P.  E. 

rating  the  concord  or  discord  of  various  combinations.  Rameau 
and  D'Alembert  about  the  same  time  decided  that  since  an  octave 
and  a  12th  (octave  plus  fifth)  were  produced  in  nature  as  the  2nd 
and  3rd  harmonics  of  a  tone,  it  was  in  the  nature  of  things  that  these 
should  be  concordant  intervals.  In  1862  Helmholtz  proposed  the 
theory  that,  with  modifications,  is  generally  held  today.  It  really 
explains,  not  why  concords  are  pleasant,  but  why  discords  are  un- 
pleasant. The  situation  appears  to  be  this :  The  normal  human  being 
gets  pleasure  from  doing  difficult  things.  The  crossword  puzzle  addict 
cares  nothing  for  simple  puzzles ;  the  more  baffling  they  are  the  more 
pleasing  it  is  to  find  a  solution.  The  sportsman  who  could  shoot  his 
partridges  on  the  ground  or  his  ducks  in  the  water  deliberately  per- 
forms the  much  more  difficult  feat  of  shooting  them  flying. 

It  appears  that  listening  to  single-note  melodies  is  too  simple,  and 
so  we  prefer  to  listen  to  complicated  tone  groups  rather  than  single 
tones.  At  the  same  time  if  a  number  of  notes  are  sounded  simul- 
taneously to  form  a  single  chord,  there  must  be  no  frequencies  intro- 
duced that  are  objectionable.  If  two  notes  do  not  have  a  simple 
numerical  ratio  between  their  frequencies,  then  either  the  funda- 
mentals or  some  harmonics  of  these  will  beat  together  to  form  harsh 
low-frequency  beat  notes,  and  the  only  notes  between  which  tolerable 
concord  exists  are  those  combinations  that  do  not  produce  such  beat 
notes. 

The  simplest  harmony  exists  between  octaves,  that  is,  between  two 
tones  where  one  is  2,  4,  8,  etc.,  times  the  other  in  frequency.  This 
was  the  only  harmony  used  by  the  Greeks.  It  is  readily  obtained  in 
singing,  since  men's  and  boys'  voices  are  about  one  octave  apart; 
in  instrumental  music  the  Greeks  obtained  it  by  inserting  a  bridge 
in  their  lyres  one-third  the  distance  along  the  strings.  To  us  this 
kind  of  harmony  sounds  childish,  flat,  uninteresting.  Later,  music 
in  Europe,  in  the  first  or  second  century  B.C.,  used  the  interval  of  the 
5th,  which  is  CG  on  our  keyboard,  a  frequency  ratio  of  3:2.  The  5th 
is  the  most  important  interval  in  modern  music,  and  stringed  instru- 
ments and  pipes  as  far  back  as  the  5th  century  B.C.  were  tuned  to 
the  intervals  C,  F,  G,  C1.  Early  in  the  Christian  era,  a  5-note  scale 
became  very  common,  both  in  Europe  and  Asia,  and  modern  Scotch 
and  Chinese  music  is  frequently  found  in  this  scale,  the  notes  on 
our  keyboard  being  C,  D,  F,  G,  A.  It  will  be  found  that  many 
traditional  melodies  are  played  using  only  these  notes,  and  that  any 
tune  restricted  to  these  notes  has  a  characteristic  quality  that  we  or- 


Mar.,  1939  ]       ELECTRICAL  PRODUCTION  OF  MUSICAL  TONES 


283 


dinarily  associate  with  Scotch  music.  This  scale  appears  to  be  very 
deeply  rooted  in  our  civilization,  almost  as  deeply  as  our  more  com- 
mon seven-note  scale. 

The  question  that  naturally  arises  is:  How  did  our  musical  scale 
develop  and  is  there  anything  unique  about  it?  If  there  were  human 
beings  on  the  planet  Mars,  for  example,  would  they  have  developed 
the  same  scale  we  use  or  would  it  be  something  quite  different?  The 
answer  is  that  they  would  probably  have  developed  our  modern  musical 
scale,  since  it  has  a  certain  uniqueness  that  no  other  sequence  of  notes 
could  have  and  is  an  attempt  to 
fit  a  musical  notation  to  the  in- 
evitableness  of  arithmetic. 

This  scale  arose  simultaneously 
in  many  parts  of  Europe  and 
Asia  over  a  period  of  some  cen- 
turies and  the  sort  of  process 
taking  place  must  have  been 
identical  in  each  locality.  There 
seems  to  be  hardly  any  doubt 
that  the  development  was  as 
follows:  A  pipe  or  string  would 
be  tuned  to  some  note,  say  C  on 
our  keyboard.  At  the  dawn  of 
musical  appreciation  the  octave 
was  added,  since  it  was  early 
found  that  octave  notes  together 
give  very  pleasing  concordance.  Later,  the  5th  was  added,  that  is, 
G,  and  we  then  had  a  three-note  scale:  C,  G,  C1.  With  the  begin- 
ning of  harmony  it  would  soon  be  realized  that  whereas  a  single- 
note  melody  could  be  made  using  any  desired  interval  between  the 
notes,  as  soon  as  two  melodies  are  combined  the  notes  in  the  scale 
have  to  be  adjusted  so  that  the  two  parts  sound  well  together. 

With  a  scale  of  C,  G,  C1,  the  next  logical  development  would  be  to 
add  a  string  or  pipe  that  would  sound  as  well  with  G  as  does  C.  This 
is  again  a  5th  above  G  and  is  D  on  our  scale.  We  now  have  four  notes 
from  which  several  harmonies  can  be  obtained:  CG,  GD,  CC1.  A 
later  development  of  harmony  also  accepted  GC1  although  this  prob- 
ably was  not  accepted  at  first.  Another  fifth  above  D  gives  A,  and 
by  following  this  process  right  up  the  scale,  we  find  that  the  notes 
shown  in  Fig.  1  are  obtained.  Going  up  by  a  5th  twelve  times  brings 


B 


FIG.  1.  The  twelve  semitones  of 
the  7-note  scale.  Each  note  has  the 
ratio  2:3  with  the  note  after  it  and 
the  ratio  3:2  with  the  note  before  it 
(based  on  relation  (3A)12  =  27). 


284  S.  T.  FISHER  [j.  s.  M.  P.  E. 

us  out  again  to  our  starting  note  C,  seven  octaves  up.  It  is  evident 
that  when  the  early  experimenters  found  they  had  again  reached  C 
they  would  consider  that  they  had  used  all  the  natural  notes  and 
would  be  content  with  this  scale  of  twelve  tones.  In  other  words, 
the  modern  musical  scale  is  an  inevitable  result  of  the  elementary 
equation  in  arithmetic: 

ON 

|  )     ==2'  (12  fifths  =  7  octaves) 

> 

129. 75  ==128 

The  equality  is  not  exact  but  is  quite  close.  The  residual  interval  is  a 
quarter  of  a  semitone,  and  is  called  the  "comma  of  Pythagoras," 
Pythagoras  having  first  pointed  out  the  mathematical  basis  of  the 
musical  scale.  A  much  more  exact  relation  is  that 

o\53 

6  i     =  231 


This  would  mean  a  53-note  octave,  which  is  obviously  too  cumbersome 
for  convenience.  With  the  12-note  scale,  if  the  inexactness  of  the 
basic  relation  is  distributed  over  the  twelve  notes,  we  have  a  scale 
that  obviously  gives  reasonably  good  concord  and  at  the  same  time 
is  fairly  convenient,  since  only  five  additional  notes  have  been  put  in 
to  fill  the  gaps  of  our  basic  structure  of  seven  notes.  It  is  interesting 
to  note  that  a  53-note  scale  was  proposed  by  Mercator,  the  map- 
maker,  about  1550,  and  about  the  year  1700  several  organs  were  con- 
structed whose  keyboard  contained  53  keys  to  the  octave.  A  subse- 
quent simplification  was  to  reduce  this  to  nineteen  notes  and  the  19- 
note  keyboard  appears  to  have  been  adopted  on  quite  a  large  scale 
during  the  Middle  Ages.  It  is  not,  however,  as  exact  as  the  12-note 
scale.  It  will  be  noticed  that  the  5-note  scale  mentioned  previously  is 
obtained  as  any  sequence  of  five  fifths.  While  from  the  foregoing  it 
is  seen  that  a  scale  of  twelve  notes  has  been  constructed,  actually, 
according  to  a  tradition  going  back  possibly  to  2000  B.C.,  a  7-note 
scale  is  used,  the  five  additional  notes  being  employed  only  as  acci- 
dentals— that  is,  for  occasional  effects — or  in  order  to  permit  music 
to  be  shifted  in  position  on  the  keyboard — that  is,  to  be  raised  or 
lowered  in  pitch.  Our  7-note  scale  is  the  scale  that  comes  as  second 
nature  to  everyone,  whether  he  has  any  musical  training  or  not; 
this  is  the  familiar  do,  re,  mi,  fa,  sol,  la,  ti,  do.  All  music  is  written  in 


Mar.,  1939]       ELECTRICAL  PRODUCTION  OF  MUSICAL  TONES  285 

this  scale,  which  in  the  key  of  C  is  represented  by  the  white  notes  on  the 
piano  keyboard.  It  is  seen  that  this  7-note  scale  goes  up  by  unequal 
increments  and,  if  we  adopt  the  usual  musical  language,  the  intervals 
between  the  notes  in  ascending  order  are  tone,  tone,  semitone,  tone, 
tone,  tone,  semitone.  When  the  black  keys  are  added  as  well,  then 
all  the  intervals  become  semitones. 

A  large  number  of  scales  have  been  constructed  on  the  7-note  basis 
with  five  additional  semitones  in  which  the  frequency  intervals  be- 
tween the  notes  have  been  arranged  according  to  a  number  of  rules 
so  as  to  give  the  nearest  approach  to  exact  harmony.  Most  of  these 
need  not  detain  us;  three,  however,  are  of  outstanding  interest: 
those  called  the  "mean- tone  scale,"  "just  scale,"  and  the  "equally 
tempered  scale."  From  the  time  of  Pythagoras  it  has  been  recognized 
that  the  mathematical  basis  of  the  musical  scale  is  inexact  and, 
Pythagoras  having  pointed  out  specifically  that  seven  octaves  are 
not  quite  equal  to  twelve  fifths,  a  number  of  schemes  have  been  de- 
vised in  the  intervening  centuries  to  distribute  this  error  in  various 
ways  over  the  scale.  The  mean-tone  scale,  proposed  by  the  Greeks 
and  used  throughout  the  Middle  Ages,  was  the  first  successful  attempt 
to  do  this.  In  this  scale  the  twelve  ascending  5ths  making  up  the 
seven  octaves  were  each  flattened  slightly  so  that  the  final  note  was 
correct  in  either  sequence.  This  scale  was  tolerable  but  suffered  from 
the  serious  defect  that  the  intervals  that  were  noticeably  inharmonious 
were  those  most  generally  used.  In  the  mean-tone  scale  a  tone  was  a 
frequency  ratio  of  9 :8,  and  a  semitone,  a  ratio  of  256 : 243 — that  is,  the 
semitone  was  not  half  the  tone:  It  will  be  seen  what  complication  this 
leads  to  when  it  is  desired  to  shift  a  piece  of  music  upward  or  down- 
ward in  the  frequency  spectrum,  the  operation  the  musician  calls 
changing  the  key.  When  this  is  done  in  the  mean-tone  scale  the 
harmony  is  badly  disturbed,  since  the  interval  of  a  fifth,  for  example, 
can  be  expanded  or  compressed,  depending  on  its  position  on  the 
keyboard.  Nevertheless,  this  scale  was  in  use  for  many  centuries  and 
was  used  by  the  earlier  of  the  great  modern  masters  of  music.  On 
keyboard  instruments  where  the  tuning  was  fairly  exact  and  where 
the  harmony  was  complicated,  it  was  very  common  for  the  musician 
to  retune  his  instrument  for  the  key  in  which  the  music  was  set.  In 
orchestral  instruments  where  the  tuning  is  much  less  exact  and  the 
source  of  origin  of  the  tones  more  diffuse,  this  was  not  done.  The 
stringed  instruments,  the  violin,  the  viola,  the  'cello,  and  the  double 
bass,  do  not  employ  frets  on  the  strings,  and  as  a  result  the  exact  into- 


286  S.  T.  FISHER  tf.  s.  M.  P.  E. 

nation  is  continually  under  control  of  the  performer,  so  that  a  violinist, 
for  example,  could  play  in  the  mean-tone  scale  and,  when  he  changed 
key,  readjust  the  intervals  so  that  they  would  still  be  harmonious 
in  the  new  key.  Factors  like  these  rendered  the  mean-tone  scale  satis- 
factory, even  to  the  great  musicians. 

About  the  year  1700,  however,  the  modern  scale  began  to  be 
adopted.  It  had  been  known  for  many  centuries  and  was  probably  in 
use  in  China  in  1000  B.C.  A  complete  description  of  it  appears  in 
Chinese  manuscripts  of  1500  A.D.,  including  the  mathematical  re- 
lations involved.  This  scale  was  proposed  and  became  adopted  be- 
cause it  has  one  great  merit:  it  permits  the  player  to  shift  from  one 
key  to  another  without  any  change  in  the  intervals  by  which  the  scale 
progresses.  There  are  twelve  intervals  in  the  musical  scale  and  the 
equally  tempered  arrangement  is  this:  that  the  frequency  ratio  of 
each  note  to  the  preceding  one  is  the  twelfth  root  of  2.  It  will  be  seen 
then  that  no  matter  what  position  on  the  keyboard  a  piece  of  music 
is  set,  the  effect  and  the  harmony  are  precisely  the  same,  the  only 
difference  being  an  overall  change  in  pitch.  The  equally  tempered 
scale  has  one  fortunate  property,  in  that  the  interval  of  a  fifth  is 
almost  exact.  It  is  likely  that  one  of  the  contributing  causes  toward 
the  adoption  of  this  scale  was  the  fact  that,  when  the  development 
of  the  piano  had  progressed  through  the  stages  of  spinet,  harpsichord, 
clavichord,  and  all  the  others,  to  the  modern  form,  the  difficulty  of 
retuning  the  instrument  for  each  key,  which  is  necessary  on  the  mean- 
tone  scale  and  without  which  the  harmonies  were  sufficiently  inac- 
curate to  be  intolerable,  made  it  imperative  to  devise  an  arrangement 
in  which  this  retuning  was  not  necessary. 

The  Greeks,  although  they  had  used  the  7-note  scale,  had  no  such 
device  as  a  change  of  key,  since  they  did  not  employ  harmony  as  we 
know  it.  They  used,  however,  a  change  of  mode,  something  that 
has  almost  disappeared  from  modern  music.  There  were  seven  modes 
in  Greek  music  and  they  were  actually  not  simple  changes  in  pitch 
of  the  whole  of  a  piece  of  music  but  a  change  from  one  sequence  of 
tones  and  semitones  to  another;  in  other  words,  there  were  seven 
different  scales.  Of  these  modes,  two  have  remained:  our  major 
mode,  which  is  the  usual  scale  C,  D,  E,  F,  G,  A,  B,  and,  our  minor 
mode  which  may,  for  example,  be  A,  B,  C,  D,  E,  F,  G.  The  Greeks 
attributed  definite  characteristics  to  each  of  their  modes  and  this 
has  been  carried  into  our  own  music,  so  much  so  that  the  expression 
"in  a  minor  key"  is  an  accepted  English  phrase. 


Mar.,  1939]       ELECTRICAL  PRODUCTION  OF  MUSICAL  TONES  287 

Another  thing  that  has  been  carried  into  modern  musical  thought 
from  the  modes  of  the  Greeks  and  more  recently  the  keys  of  the  mean- 
tone  scale,  is  the  fiction  that  many  musicians  believe  that  different 
keys  on  the  equally  tempered  scale  have  characteristic  qualities.  This 
simply  is  not  so.  A  change  of  key  in  the  equally  tempered  scale  re- 
sults in  nothing  except  a  change  of  pitch.  There  is  no  change  in  the 
character  of  the  music.  This  can  be  demonstrated  beyond  any  ques- 
tion when  it  is  pointed  out  that  the  key  of  F*  major  (6  sharps)  uses 
the  identical  notes  of  the  key  of  Gb  major  (6  flats).  Incidental  effects 
do,  however,  exist.  A  violinist,  unaccompanied,  tends  to  play  in 
true  harmonic  intervals;  a  change  of  key  changes  the  number  of 
strings  that  are  played  "open,"  i.  e.,  at  their  maximum  length.  Some 
instruments  have  different  qualities  for  different  positions  of  playing, 
as  the  piano,  and  in  some  instruments  some  notes  may  be  poor,  as 
C*  on  the  flute,  and  some  notes  may  be  unusually  good,  as  high  D  on 
the  same  instrument. 

The  Chinese  musical  scale  of  which  mention  has  been  made  before, 
consists  of  our  12-note  equally  tempered  scale,  but  the  music  has 
this  fundamental  difference:  that  all  twelve  notes  are  employed  in 
any  given  piece  of  music — that  is,  this  music,  as  a  musician  would 
say,  belongs  in  the  chromatic  scale.  This  scale  was  arrived  at  by 
the  Chinese  as  an  outcrop  of  their  study  of  numerology  and  is  pre- 
cisely correct.  They  took  a  bamboo  flute  tube  of  a  given  length  and 
made  up  the  scale  by  shortening  the  tube  in  twelve  steps,  the  decre- 
ment in  length  of  each  step  being  a  fixed  proportion  of  the  length  of 
the  preceding  step. 

The  equally  tempered  scale  was  not  much  used  in  England  until 
about  the  year  1850  and  was  not  generally  used  for  some  years  after- 
ward. At  the  Exhibition  of  1851,  for  example,  it  is  said  that  not  a 
single  organ  exhibited  was  in  the  equally  tempered  scale  but  that  all 
were  in  the  mean-tone  scale.  This  seems  to  be  rather  a  severe  com- 
mentary on  the  English  ear  for  harmony,  since  the  equally  tempered 
scale  had  been  adopted  generally  in  Europe  one  hundred  and  fifty 
years  before,  solely  on  the  grounds  that  the  mean-tone  scale  was  in- 
tolerable except  in  the  key  of  C. 

The  other  scale  of  fundamental  interest  in  music  is  the  just,  or 
harmonic,  scale.  This  is  a  theoretically  correct  scale  but  suffers  from 
the  disability  that  it  can  be  played  only  in  one  key.  It  consists  of  a 
series  of  notes  whose  frequencies  are  represented  by  the  numbers  1, 
Ys,  6A>  Vs,  3A,  Vs,  15A,  and  2.  It  will  be  seen  that  all  these  notes 


288 


S.  T.  FISHER 


[J.  S.  M.  P.  E. 


bear  simple  numerical  relations  to  the  key  note  and  therefore  this 
scale  is  very  rich  in  exact  harmonies.  Unfortunately  there  occur  in 
it  three  different  sizes  of  intervals,  one  having  a  ratio  of  9/8,  one 
having  a  ratio  of  10/9,  and  one  having  a  ratio  of  15/16,  so  that  at 
present  no  keyboard  instrument  is  tuned  to  this  scale.  There  seems 
very  little  doubt,  however,  that  when  a  violinist  is  playing  unaccom- 
panied by  a  keyboard  instrument,  he  plays  in  this  scale,  and  measure- 
ments made  by  Helmholtz  and  others  indicate  that  this  is  so.  In 
Table  I  is  shown  a  comparison  of  the  harmonic  or  just  scale  and  the 
tempered  scale  intervals.  This  table  shows  two  things:  first,  the 

TABLE  I 

Comparison  of  Harmonic  and  Tempered  Scale  Intervals 


Note 

Tem- 
pered 
Scale 
Fre- 
quency 

Har- 
monic 
Scale 
Fre- 
quency 

Harmonic  Scale:  Harmonics  of 

C 

G 

D 

F 

A 

c 

1000 

1000 

1000 

1031 

985 

1000 

1042 

c# 

1059 

D 

1125 

1125 

1125 

1125 

1125 

1146 

Df 

1189 

E 

1260 

1250 

1250 

1266 

1250 

F 

1355 

1333 

1375 

1312 

.•; 

1333 

Ft 

1414 

G 

1498 

1500 

1500 

1500 

1500 

1458 

G# 

1587 

A 

1682 

1667 

.  . 

1687 

1687 

1667 

1667 

At 

1782 

B 

1888 

1875 

. 

1875 

.  . 

. 

1875 

C 

2000 

2000 

2000 

2062 

1969 

2000 

2083 

divergence  of  the  tempered  scale  from  a  theoretically  correct  scale, 
and,  second  the  excellence  of  the  harmonies  occurring  in  the  harmonic 
scale  between  fundamentals  and  harmonics.  In  the  tempered  scale 
no  such  fortunate  relations  exist,  since  the  fundamental  intervals 
are  all  different  from  the  theoretical  values.  The  harmonics  are  more 
and  more  divergent  in  notes  occurring  higher  on  the  keyboard  as  we 
go  up  in  frequency. 

I  have  never  heard  a  keyboard  instrument  played  in  the  just  scale 
but  I  think  there  is  little  doubt  that  it  would  give  definitely  a  more 
pleasing  effect  than  our  usual  tempered-scale  instrument.  This  would 
be  particularly  true  on  the  organ  where  extremely  complicated 


Mar.,  1939  ]       ELECTRICAL  PRODUCTION  OF  MUSICAL  TONES 


289 


harmonies  both  of  fundamental  and  harmonics  are  obtained.  Helm- 
holtz  in  1860  commented  on  this  and  spoke  of  the  extreme  pleasure  he 
got  from  playing  a  justly  tuned  instrument  after  playing  on  a  tem- 
pered-scale  piano.  Of  an  organ  tuned  in  the  tempered  scale  he  said, 
"When  the  mixture  stops  are  played  in  full  chords,  a  hellish  row  must 
ensue  and  organists  must  submit  to  their  fate."  We  still  submit  to 
our  fate.  The  just  scale  gives  almost  exact  harmonies,  but  can  not  be 
played  in  more  than  one  key  without  re  tuning ;  and  the  tempered  scale 

TABLE  n 

Comparison  of  Scale  of  Equal  Temperament  with  Hammond  "Tempered  Harmonic" 

Scale 


Note 

Tem- 
pered 
Scale 
Fre- 
quencies 

Frequencies  of  Harmonics 
of  Notes  in  Lower  Octaves  Which 
Appear  in  This  Octave 

2nd,  4th,  8th 

3rd,  6th 

5th 

7th 

Tern- 
Natural    pered 

Tern- 
Natural    pered 

Tern- 
Natural       pered 

Tern- 
Natural    pered 

C 

1000 

1000      1000 

1001      1000 

992       1000 

985       1000 

c# 

1059 

1059     1059 

1061     1059 

1051       1059 

1040       1059 

D 

1125 

1125     1125 

1124     1125 

1114       1125 

1102       1125 

D* 

1189 

1189     1189 

1190     1189 

1180       1189 

1168       1189 

E 

1260 

1260     1260 

1262     1260 

1250       1260 

1237       1260 

F 

1335 

1335     1335 

1337     1335 

1324       1335 

1311       1335 

F# 

1414 

1414     1414 

1416     1414 

1406       1414 

1389       1414 

G 

1498 

1498     1498 

1500     1498 

1486       1498 

1472       1498 

.G* 

1587 

1587     1587 

1589     1587 

1575       1587 

1559       1587 

A 

1682 

1682     1682 

1688     1682 

1669       1682 

1652       1682 

A* 

1782 

1782     1782 

1784     1782 

1767       1782 

1750       1782 

B 

1888 

1888     1882 

1890     1882 

1875       1882 

1853       1882 

C 

2000 

2000    2000 

2003     2000 

1984      2000 

1969      2000 

can  be  played  in  all  keys,  but  what  should  be  harmonious  relations  be- 
tween notes  in  the  same  octave  are  actually  discords.  From  Table 
II  can  be  seen  an  even  more  important  source  of  discord,  particu- 
larly noticeable,  as  it  was  to  Helmholtz,  in  organs.  This  is  the  fact 
that  the  3rd,  5th,  6th,  7th,  10th,  and  some  of  the  higher  harmonics 
of  notes  at  the  lower  end  of  the  keyboard  do  not  duplicate  notes  on 
the  upper  end  of  the  keyboard,  but  are  noticeably  out  of  tune  with 
them.  In  the  orchestra  then,  the  3rd,  5th,  6th,  and  7th  harmonics  of 
the  string  basses  must  cause  discord  with  the  violins  and  woodwind; 
but  the  effect  is  not  particularly  noticeable,  first,  because  we  are  used 
to  it,  second,  because  the  instruments  are  all  slightly  out  of  tune  in 


290  S.  T.  FISHER  [J.  S.  M.  P.  E. 

various  ways,  and  third,  because  the  source  of  the  discordant  sounds  is 
spread  over  a  relatively  large  angle  at  the  listener's  ear.  In  the  pipe- 
organ,  the  situation  is  definitely  worse  when  the  mixture  stops  are 
played.  Mixture  stops  are  those  in  which  rows  of  pipes  representing 
2nd,  3rd,  4th,  and  even  up  to  the  10th  harmonic  are  coupled  to  the 
fundamental  pipes  being  played;  so  clash  is  inevitable  between  the 
natural  harmonics  of  the  fundamental  notes,  which  in  many  cases 
are  very  strong,  and  these  synthetic  harmonics,  which  lie  strictly 
in  the  tempered  scale.  In  the  piano,  the  discord  between  the  nat- 
ural 7th  harmonic  and  the  tempered-scale  notes  was  early  recognized 
as  disagreeable,  and  today  all  pianos  are  so  arranged  that  the  7th 
harmonic  is  largely  suppressed ;  this  is  done  also  in  organs  and  in  the 
brass  wind  instruments.  The  oboe,  among  the  orchestra  instruments, 
is  characterized  by  a  strong  7th  harmonic,  and  its  harsh,  penetrating 
quality  may  be  largely  due  to  the  discords  thereby  produced.  Two 
possibilities  of  remedying  this  situation  theoretically  are  not  open 
practically  to  the  constructor  of  traditional  musical  instruments.  The 
theoretical  possibilities  are :  First,  to  supress  all  natural  harmonics  and 
use  only  tempered  ones.  This  is  obviously  not  possible  practically, 
except  to  some  extent  in  the  case  of  the  pipe-organ.  Second,  to  shift 
to  the  just  scale,  so  that  in  most  important  instances  the  natural 
harmonics  appear  almost  exactly  on  the  scale.  This  results  in  the 
limitation  of  the  instrument  to  a  single  key. 

By  far  the  best  answer  to  date  has  been  supplied  by  Laurens  Ham- 
mond in  his  electric  organ.  In  this  instrument,  musical  qualities  are 
synthesized  from  pure  sine  waves,  and  all  the  frequencies  used  in 
the  synthesis  lie  on  the  tempered  scale.  In  other  words,  natural  har- 
monics are  entirely  suppressed  and  tempered  harmonics  substituted. 
In  no  other  instrument,  to  my  knowledge,  is  this  done,  and  while 
the  results  are  not  immediately  perceptible  to  the  lay  ear,  the  char- 
acteristically harmonious  effect  of  the  Hammond  organ  that  becomes 
apparent  after  some  familiarity  with  it  must  be  ascribed  to  this  basic 
improvement. 

A  topic  of  great  importance  that  we  shall  consider  briefly  is  that  of 
summation  and  difference  tones.  Communication  engineers  are 
familiar  with  the  effect  of  superimposing  two  different  frequencies 
and  transmitting  them  through  a  non-linear  network.  We  ordinarily 
say  that  side-bands  are  produced,  consisting  of  sum  and  difference 
frequencies.  The  ear  is  non-linear  to  a  marked  extent,  with  the 
result  that  when  two  notes  are  sounded  loudly,  there  become  per- 


Mar.,  1939]       ELECTRICAL  PRODUCTION  OF  MUSICAL  TONES  291 

ceptible  a  number  of  other  tones  bearing  related  frequencies.  The 
most  prominent  are  the  2nd  harmonics  and  the  sum  and  difference 
frequencies  of  the  two  notes.  It  can  be  shown  that  if  three  tones  repre- 
sented in  frequency  by  the  numbers  4,  5,  and  6  are  sounded,  the  hu- 
man ear  will  respond  as  if  every  frequency  from  1  to  18  were  present. 
It  can  be  readily  demonstrated  with  an  organ,  or  for  that  matter 
with  two  oscillators,  that  if  two  tones  of  frequencies  2  and  3  are 
sounded  together,  a  note  of  frequency  1  is  plainly  audible.  This  is 
the  phenomenon  that  lets  us  hear  the  fundamental  tones  of  a  man's 
voice  over  the  telephone,  although  these  tones  go  down  to  90  cycles 
and  the  telephone  transmits  little  or  nothing  below  300  cycles.  It 
lets  us  hear  the  low  notes  of  the  piano,  down  to  28  cycles,  although 
acoustically  most  pianos  are  quite  incapable  of  radiating  a  perceptible 
amount  of  power  at  these  frequencies.  This  phenomenon,  and  par- 
ticularly the  formation  of  difference  tones,  is  very  important  in  musi- 
cal instruments.  In  the  pipe-organ  where  it  is  particularly  notice- 
able, it  is  called  "acoustic  bass."  The  summation  tones,  especially, 
give  rise  to  serious  discords.  Suppose  we  have  a  note  sounded  with 
strong  3rd  and  4th  harmonics.  The  difference  tone  of  the  harmonics 
is  simply  a  strengthened  fundamental,  but  the  summation  tone  is 
the  7th  harmonic,  which  is  extremely  discordant  in  the  tempered  scale. 
In  the  case  of  drums,  bells,  and  other  instruments  in  which  the  par- 
tials  are  not  harmonics,  difference  tones  can  not  be  relied  on  to  supply 
a  bass  that  is  not  transmitted. 

Musical  tones,  being  sounds,  are  always  produced  by  vibrating 
mechanical  systems,  capable  of  acoustic  radiation.  The  piano,  the 
violin,  the  pipe-organ,  or  the  human  voice  are  examples  in  which  the 
radiating  element  is  actuated  mechanically.  In  all  these  instances, 
the  further  feature  exists  that  the  radiating  element  is  also  the  gen- 
erator. We  can  conceive  of  a  variety  of  ways  in  which  a  piano,  for 
example,  could  be  modified  by  the  application  of  electricity.  The 
striking  mechanism  could  be  actuated  by  electromagnets  instead  of 
directly  by  the  performer;  the  strings  could  be  sounded  by  a  micro- 
phone-hummer arrangement;  the  strings  could  be  enclosed  and  the 
sound  picked  up  by  a  microphone  and  reproduced  through  an  ampli- 
fier and  loud  speaker;  or  instead  of  a  microphone,  a  direct  electro- 
magnetic or  electrostatic  coupling  to  the  strings  could  be  used.  While 
all  these  schemes  and  combinations  involve  electricity  in  the  produc- 
tion of  musical  tones,  we  are  most  concerned  with  the  case  in  which 
an  electric  current  of  the  required  character  is  generated;  and  when 


292  S.  T.  FISHER  [j.  s.  M.  P.  E. 

this  current,  after  amplification,  is  passed  through  a  loud  speaker,  a 
musical  tone  is  produced. 

There  are  two  fundamentally  different  ways  in  which  such  an  elec- 
tronic instrument  may  be  arranged  to  produce  musical  tones:  The 
tone  can  be  generated  in  the  first  instance  as  a  complete  wave  of  the 
desired  character,  which  is  then  amplified  and  reproduced;  or  it  can 
be  produced  as  the  addition  of  a  group  of  sine- wave  frequencies  lying 
in  a  harmonic  series. 

In  the  Hammond  organ  the  latter  arrangement  is  used.  The  tone- 
generator  consists  of  ninety-one  miniature  tone- wheels.  These  wheels 
are  made  of  steel,  and  each  has  a  sine  wave  cut  in  the  periphery. 
A  single  permanent-magnet  pole-piece  with  a  coil  wound  on  it  is  used 
with  each  generator.  All  the  generators  are  driven  from  a  single 
synchronous  motor  with  elaborate  precautions  taken  to  prevent  fre- 
quency-modulation flutter.  These  consist  of  a  damped  low-pass 
mechanical  filter  in  the  main  drive,  and  of  a  similar  small  section  in 
the  drive  to  each  pair  of  tone- wheels.  The  generator  assembly  con- 
sists of  two  groups  of  jack  shafts,  on  each  of  which  is  mounted  two 
tone-wheels,  separated  in  frequency  by  an  integral  number  of  octaves, 
and  a  gear  and  mechanical  low-pass  filter.  These  two  groups  are  driven 
by  gears  mounted  on  a  main  drive-shaft  and  one  group  lies  on  each 
side  of  it.  It  is  evident  that  the  limitations  of  motor  speed,  and  of 
cutting  whole  numbers  of  teeth  on  gears  and  tone- wheels,  will  not  per- 
mit an  exactly  precise  duplication  of  the  tempered  scale.  Com- 
promises have  been  necessary,  but  they  are  of  negligible  importance. 
This  could  not  be  said  of  such  departures  from  the  just  scale,  if  it 
were  in  use  instead  of  the  equally  tempered  scale.  Note  this  point, 
however:  if  the  just  scale  were  used  for  this  organ,  no  departures 
would  be  necessary,  since  all  the  frequencies  bear  simple  fractional 
relations  to  each  other.  In  the  Hammond  organ,  the  octave  intervals 
are  maintained  precisely,  this  being  the  purpose  of  the  small  jack 
shafts,  each  with  two  tone-wheels  mounted  on  it. 

The  output  of  each  pick-up  coil  is  passed  through  a  low-pass  filter 
to  reduce  it  as  nearly  as  possible  to  a  sine  wave.  In  this  instrument, 
provision  is  made  to  synthesize  tones  using  a  fundamental  or  1st 
harmonic,  the  2nd,  3rd,  4th,  5th,  6th,  and  8th  harmonics;  the  sub- 
harmonic,  or  octave  below  the  fundamental;  and  the  sub-third,  a 
fifth  above  the  fundamental.  From  these  nine  partials  pipe-organ 
voices  can  be  imitated,  in  some  cases  perfectly,  in  all  cases  adequately, 
and  this  is  true  of  most  of  the  orchestra  instruments.  Obviously, 


Mar.,  1939]      ELECTRICAL  PRODUCTION  OF  MUSICAL  TONES  293 

tones  can  be  set  up  that  are  entirely  new  qualities,  not  produced  by 
traditional  instruments.  It  is  evident  that  each  generator  output 
appears  in  a  number  of  places  on  the  keyboard.  For  instance,  if  we 
have  a  generator  that  is  the  fundamental  frequency  of  Middle  C,  it  is 
the  2nd  harmonic  of  the  C  below,  the  4th  harmonic  of  the  C  below 
that,  and  the  8th  harmonic  of  the  C  below  that  again.  It  is  the  sub- 
harmonic  of  the  C  above,  the  3rd  harmonic  of  the  F  an  octave  and  a 
fifth  below,  and  the  6th  harmonic  of  the  F  two  octaves  and  a  fifth 
below.  It  is  the  sub-third  harmonic  of  the  F  below  and  the  5th  har- 
monic of  the  G#  two  octaves  and  a  third  below. 

Since  a  tone  produced  by  the  organ  may  include  all  nine  partials, 
each  key  carries  an  assembly  of  nine  contact  springs.  The  moving 
sides  of  these  contacts  connect  to  the  generators,  the  stationary  sides 
through  the  switching  mechanism  that  selects  the  desired  harmonics 
and  adjusts  their  relative  strengths  to  an  amplifier  contained  in  the 
console.  The  output  from  this  amplifier  is  fed  to  as  many  power- 
amplifier  tone-cabinet  units  as  may  be  necessary. 

The  relative  amplitudes  of  the  generator  outputs  are  adjusted  by 
sliding  the  pole-pieces.  The  output  is  adjusted  to  a  curve  rising 
steeply  from  the  high-frequency  end  to  the  low-frequency  end.  Aside 
from  the  harmonic-suppression  filters  on  each  generator,  no  electrical 
filters  are  used.  From  the  previous  discussion  of  the  tempered  scale, 
it  will  be  realized  that  the  suppression  of  natural  harmonics  is  of  the 
greatest  importance  except  in  the  top  octave  of  the  keyboard,  where 
they  fall  outside  the  range  of  any  of  the  generator  fundamental  fre- 
quencies. Accordingly  no  harmonic  filters  are  employed  on  the  top 
octave  of  generators,  to  permit  the  added  brilliance  of  an  extended 
harmonic  range. 

The  strength  of  each  harmonic  in  a  tone  can  be  set  in  eight  steps 
of  3  db.  each,  or  cut  out  entirely.  Controls  are  provided  so  that  four 
tones  can  be  set  up,  two  for  each  manual  or  keyboard,  and  can  be 
brought  in  by  pressing  a  button.  In  addition,  a  limited  range  of 
tones  can  be  set  for  the  pedals.  Eighteen  of  the  usual  pipe-organ 
qualities  are  permanently  adjusted,  and  nine  may  be  used  on  each 
manual  by  pressing  the  appropriate  switches  at  the  left  of  the  key- 
board. The  volume-control,  operated  by  a  foot-pedal,  has  a  total 
range  of  30  db. 

A  tremulant  is  provided  by  a  motor-driven  potentiometer,  and  the 
effect  is  variable  by  means  of  a  manually  operated  potentiometer 
connected  across  it.  An  effect  of  great  importance  that  is  provided 


294  S.  T.  FISHER  [j.  s.  M.  P.  E. 

is  the  so-called  "chorus  effect."  It  is  this  that  enables  us  to  tell  twenty 
violins  playing  in  unison  from  one  violin  playing  loudly,  and  is  par- 
ticularly noticeable  in  the  pipe-organ,  where  frequently  a  great  num- 
ber of  pipes  of  the  same  pitch  may  be  sounded  together.  With  a 
number  of  separate  sources,  it  is  inevitable  that  small  frequency  differ- 
ences should  occur  and  this,  together  with  random  and  changing 
phase  relations,  is  imitated  in  Hammond's  organ  by  having,  for 
each  generator  representing  a  keyboard  note,  another  generator 
slightly  out  of  tune  with  it,  whose  output  is  quite  low.  This  second 
set  of  generators  can  be  connected  at  will  by  a  control  at  the  right 
of  the  keyboard. 

The  Hammond  organ  is  widely  accepted  by  musicians  and  is  now 
in  general  use  throughout  the  world.  Among  other  electronic  instru- 
ments that  have  been  commercially  exploited  is  the  Everett  Orgatron, 
an  instrument  that  employs  wind-blown  reeds  with  electrostatic  pick- 
ups. The  reeds  are  in  a  sound-proof  chamber,  and  different  tone 
qualities  are  obtained  by  using  different  banks  of  reeds  and  by  shap- 
ing the  response  curve  of  the  amplifier.  This  instrument  gives  the 
characteristic  reed-organ  quality.  There  are  several  pianos  with 
electrostatic  or  electromagnetic  pick-ups  on  the  strings  that  are 
said  to  be  extremely  effective,  since  they  permit  the  elimination  of 
the  sounding  board,  the  bass  can  be  enhanced  at  will,  and  a  wide 
volume  range  can  be  obtained  without  the  necessity  of  tremendous 
exertion  by  the  performer  or  of  changing  the  quality  of  the  sound 
produced. 

A  number  of  organs  using  vacuum-tube  oscillators  have  been  pro- 
duced commercially,  some  using  a  separate  oscillator  for  each  note, 
some  using  only  twelve  oscillators,  one  for  each  note  in  the  lowest  oc- 
tave, all  the  upper  notes  being  obtained  as  harmonics.  There  appear 
to  be  two  objections  to  these  instruments :  first,  the  large  number 
of  vacuum-tubes  involved — in  the  hundreds — and  second,  the  diffi- 
culty of  keeping  them  exactly  in  tune.  These  problems  are  undoubt- 
edly capable  of  solution,  however. 

An  outstanding  example  of  the  principle  of  the  generation  directly 
of  a  complex  tone,  instead  of  synthesis  from  harmonic  components, 
is  the  Robb  Wave-Organ,  designed  by  Morse  Robb  and  constructed 
at  Belleville,  Ontario.  This  interesting  and  successful  instrument 
consists  of  a  console  connected  by  a  multi-conductor  cable  to  the  tone- 
generating  unit,  and  the  usual  amplifiers  and  loud  speakers.  The 
generator  consists  of  twelve  spindles,  one  for  each  note  in  the  octave, 


Mar.,  1939 ]       ELECTRICAL  PRODUCTION  OF  MUSICAL  TONES  295 

each  with  a  number  of  disks,  and  driven  through  belts  and  pulleys 
from  a  single  motor.  On  these  disks  are  cut  the  complex  wave-forms 
it  is  desired  to  reproduce,  so  that,  subject  only  to  the  limitations  of 
the  amplifier  and  loud  speaker  equipment,  almost  any  tone  can  be 
accurately  reproduced.  A  very  large  number  of  generator  disks  is 
required — one  for  each  keyboard  note,  for  each  tone  color.  The  total 
number  is  substantially  reduced,  however,  by  using  generators  in 
the  form  of  cylinders,  not  flat  disks;  a  number  of  pick-ups  are  em- 
ployed on  each  cylinder,  at  different  points  in  the  height  of  the  wave- 
form, and  different  pick-ups  and  different  combinations  of  pick-ups 
give  wide  variations  in  tone-quality.  This  organ  is  provided  with  a 
group  of  standard  pipe-organ  stops,  the  disks  being  cut  from  oscillo- 
grams  made  from  an  acoustic  pick-up  from  a  pipe-organ.  The  Robb 
Wave-Organ  embodies  a  number  of  ingenious  points  in  its  design  and 
must  be  regarded  as  a  serious  musical  instrument. 

The  classical  work  on  electrical  musical  instruments  was  carried 
out  by  Thaddeus  Cahill,  between  the  years  1895  and  1905.  All 
the  modern  inventors  have  drawn  largely  on  his  work,  which  is  ex- 
plained and  described  exhaustively  in  five  patent  applications  drawn 
up  by  him.  Cahill  appears  to  have  been  the  first  man  to  conceive 
the  idea  of  the  electrical  production  of  musical  tones.  He  produced 
several  models  of  an  instrument  which  he  called  the  Telharmonium. 
To  ship  one  of  these  instruments  from  his  laboratory  at  Holyoke, 
Mass.,  to  New  York  required  forty  railway  cars;  the  instrument 
weighed  over  200  tons  and  cost  upward  of  a  quarter  of  a  million  dol- 
lars! Cahill  intended  to  use  his  Telharmonium  to  transmit  music 
over  telephone  lines  to  subscribers;  the  plan  finally  fell  through  be- 
cause of  cross-talk  into  adjacent  circuits. 

Cahill  clearly  understood  all  the  theoretical  and  practical  problems 
involved  in  the  electric  organ;  he  outlines  them  in  great  detail,  to- 
gether with  the  extraordinarily  ingenious  features  of  his  Telharmo- 
nium. Since,  of  course,  he  had  neither  vacuum-tube  amplifiers  nor 
modern  loud  speakers,  he  had  to  generate  relatively  great  amounts  of 
power.  He  employed  generators  in  the  form  of  wound-rotor  alterna- 
tors, commutators  (he  used  the  word  "rheo tomes")  and  vibrators. 
The  bed-plate  of  his  main  generator  assembly  was  sixty  feet  long.  He 
understood  and  used  low-pass  and  band-pass  filters  and  matching  net- 
works of  many  configurations,  and  understood  the  impedance  rela- 
tions of  his  circuits  and  the  importance  of  satisfactory  transient  char- 
acteristics, all  these  at  a  time  when  such  knowledge  was  not  general 


296  S.  T.  FISHER  [j.  s.  M.  P.  E. 

in  any  branch  of  communication  engineering.  Initially  he  used  great 
multiple  loud  speakers  made  of  vibrating  magnets  clamped  to  hard- 
wood bars;  later  he  used  telephone  receivers  with  conical  horns.  His 
Telharmonium  provided  much  more  complete  facilities  to  the  per- 
former than  has  any  electrical  or  acoustic  organ  before  or  since.  He 
employed  the  methods  both  of  tone  synthesis  from  pure  tones,  and  of 
the  generation  of  complex  tones.  An  article  in  1906  in  The  Electrician 
describes  the  performance  of  the  Telharmonium : 

It  is  evident  that  the  constructional  features  of  the  electrical  mechanism  are 
exceedingly  elaborate.  It  is  believed,  however,  that  the  results  obtained  fully 
justify  the  means  employed.  There  can  be  no  doubt  as  to  the  absolute  accuracy 
of  the  relative  pitches  of  the  various  notes  produced,  nor  as  to  the  beauty  and 
purity  of  the  resultant  music.  Although  the  horn  of  the  receiver  resembles  that 
of  a  phonograph,  there  is  nothing  about  the  music  itself  to  suggest  the  phonograph, 
the  harsh  sounds  and  disagreeable  overtones  of  which  are  entirely  lacking.  The 
quality  of  the  sound  is  pure  and  sweet  and  the  volume  is  such  that  the  largest 
known  auditorium  can  be  served  without  the  use  of  an  excessive  number  of  re- 
ceivers, while  the  character  and  expression  of  the  music  is  under  the  control  of 
the  musician  to  an  extent  not  previously  reached  in  any  musical  instrument. 

The  question  will  naturally  arise  in  the  mind  of  the  reader  as  to  the  practical 
use  of  this  complicated  and  expensive  apparatus.  The  plans  of  the  inventor  are 
to  distribute  music  from  a  central  station  to  hotels,  restaurants,  theaters,  and 
private  homes.  The  remarkable  purity  and  strength  of  the  sounds  produced 
electrically,  enabling  a  few  performers  at  a  central  station  to  produce  orchestral 
music  at  a  thousand  places,  strikes  the  imagination  and  it  seems  not  improbable 
that  at  no  distant  day  orchestral  music  for  the  dinner  table  will  be  as  common  in 
the  homes  of  the  people  as  it  is  now  in  the  great  hotels.  Music  of  different  sorts 
during  the  evening,  and  slumber  music  during  the  small  hours  of  the  night  com- 
ing to  the  listener  by  electricity  from  a  central  station,  seem  likely  in  a  few  months 
to  be  accomplished  facts  in  one  or  more  American  cities. 

The  design  of  the  transmission  circuits  of  an  electrical  organ  such 
as  the  Hammond  presents  a  number  of  problems.  The  frequency 
range  is  from  32  to  above  8000  cycles,  and  the  maximum  output  oc- 
curs at  the  lowest  frequency.  This  involves  the  design  of  amplifiers 
and  loud  speakers  whose  efficiency  and  power-handling  capacity  are 
unimpaired  at  32  cycles.  From  the  discussion  of  the  tempered  scale, 
it  will  be  realized  that  the  suppression  of  natural  harmonics  in  the 
amplifier-loud  speaker  system  is  of  the  greatest  importance.  This 
requires  careful  design  of  the  amplifier,  and  of  the  acoustic  loading  on 
the  loud  speaker  so  that  even  at  the  lowest  frequencies  no  break- 
up occurs. 

Another  point  largely  ignored  in  ordinary  transmission  systems,  of 
great  importance  in  a  system  for  the  transmission  of  organ  tones 


Mar.,  1939  ]       ELECTRICAL  PRODUCTION  OF  MUSICAL  TONES 


297 


is  that  in  a  wave  made  up  of  a  harmonic  series  with  random  phase 
relations,  the  peak  value  of  the  wave  reaches  a  relatively  high  value 
for  a  given  rms.  value.  In  other  words,  the  power  output  capacity 
of  the  amplifier  is  largely  reduced  due  to  the  wave-form  of  the  tone 
being  transmitted.  Fig.  2  shows  the  magnitude  of  this  effect. 

What  significance  can  engineers  draw  from  the  facts  that  have  been 
presented?  What  of  the  future  development  of  the  instrumentalities 
of  music  ?  One  possibility  that  is  of  overwhelming  importance  now  ap- 
pears: that  is,  to  abandon  the  tempered  scale  in  favor  of  the  just 
scale.  It  has  been  seen  what  inaccuracies  of  harmony  are  inherent 


o- 

?T 

O    —  c 

r; 

- 

s  * 

-9- 
-0 

"i 

1 

1 

~| 

-| 

s      4      ft      %     r  i      6      m      to 

N»-  OF  EQUAL  HAJZNVONIC    FREQUEIHCY    COMPONENTS 

FIG.  2.     Decrease  of  maximum  output  of  amplifiers 
for  wave-form  consisting  of  equal  harmonic  components. 

in  the  tempered  scale.  Its  main  virtue  is  its  convenience,  the  fa- 
cility it  provides  for  changing  key.  The  just  scale  must  be  retuned 
for  each  key,  but  is  the  ideal  solution  to  the  7-note  scale  since  it  pro- 
vides the  largest  number  of  exact  harmonies  possible.  This  makes  its 
use  impossible  in  the  traditional  keyboard  instruments.  The  stringed 
instruments  in  the  orchestra  can  readily  play  the  just  scale  in  any 
key,  since  the  pitch  can  be  continuously  varied ;  the  wind  instruments 
can  also,  by  means  of  changing  the  length  of  the  air  column  with  tele- 
scoping joints  for  small  changes  of  pitch  and  by  using  different  instru- 
ments for  large  changes.  With  the  advent  of  electrical  keyboard 
instruments  a  new  era  in  music  is  initiated.  It  is  now  possible  to  tune 
such  an  instrument  in  the  just  scale,  and  to  change  key  electrically 
by  altering  the  frequency  of  the  generating  devices.  The  implica- 
tions are  too  broad  to  be  touched  upon  in  this  paper,  but  from 
what  has  been  said  it  will  be  apparent  that  the  resulting  improve- 
ment in  the  harmoniousness  of  our  music  would  be  large. 


A  COLOR-TEMPERATURE  METER* 
E.  M.  LOWRY  AND  K.  S.  WEAVER** 


Summary. — The  recent  advances  in  color  photography  have  made  more  apparent 
than  ever  before  the  need  for  some  simple  and  accurate  method  for  the  estimation  of 
the  color -temperature  of  light-sources.  Photographers,  whether  professional  or  ama- 
teur, are  only  too  well  aware  of  the  influence  which  the  quality  of  the  illumination  has 
on  the  color  rendering  of  photographic  subjects.  For  example,  the  difference  in  color- 
temperature  between  general-purpose  tungsten  filament  lamps,  and  studio  modeling 
lamps,  or  between  modeling  lamps  and  photoflood  lamps,  is  often  the  deciding  factor 
between  correct  and  incorrect  photographic  color  reproduction.  In  order  that  the 
photographer  may  easily  determine  the  quality  of  lighting  which  he  is  using  and  make 
the  proper  adjustments  to  secure  standard  lighting  conditions,  an  instrument  that  is 
at  once  compact,  simple  in  operation,  and  accurate,  has  been  developed  in  these 
laboratories.  No  auxiliary  light-source  is  required  for  making  measurements  since 
each  source  is  tested  by  means  of  the  radiant  energy  which  it  itself  emits.  In  this 
paper  a  discussion  of  the  principles  applied  in  construction  of  the  instrument,  a  de- 
scription of  the  instrument,  and  data  showing  the  probable  error  of  results  are  given. 

The  recent  advances  in  color  photography  have  made  more  appar- 
ent than  ever  before  the  need  for  some  simple  and  accurate  method 
for  determining  the  characteristics  of  light-sources.  If  the  spectrum 
of  a  source  such  as  a  tungsten  lamp  is  examined  and  the  relative 
energy  of  light  of  each  wavelength  determined  and  plotted  against 
the  wavelength,  a  graph  will  be  obtained  known  as  a  spectral  energy 
distribution  curve.  Fig.  1  is  such  a  graph  for  a  100-watt  tungsten 
lamp.  It  has  been  found  that  the  spectral  distribution  of  energy 
throughout  the  spectrum  of  an  incandescent  solid,  for  example,  the 
filament  of  a  tungsten  lamp,  is  practically  independent  of  the  material 
of  which  it  is  composed  and  is  dependent  only  upon  the  temperature. 
As  the  temperature  of  an  incandescent  solid  is  raised,  the  wave- 
length at  which  the  maximum  energy  is  emitted  moves  from  the 
long  toward  the  short-wave  region  of  the  spectrum.  This  shift 
of  the  maximum  with  temperature,  illustrated  in  Table  I,  gives  rise 

*Presented  at  the  1938  Fall  Meeting  at  Detroit,  Mich. ;  received  October  28, 
1938.    Communication  No.  698  from  the  Kodak  Research  Laboratories. 
**Eastman  Kodak  Company,  Rochester,  N.  Y. 

298 


A  COLOR-TEMPERATURE  METER 


299 


to  the  common  experience  that  as  the  glowing  material  becomes 
hotter  it  appears  whiter.  Since  the  most  efficient  thermal  radiator 
is  a  completely  black  body,  the  energy  distribution  may  be  said  to 
correspond  to  that  of  a  "black  body"  at  a  given  temperature.  The 


WAVG   UCNttTH   IN 


FIG.  1.  Graph  showing  relative  spec- 
tral distribution  of  energy  for  a  100-watt 
tungsten  lamp. 

temperature  is  measured  from  the  absolute  zero  on  the  Centigrade 
scale  stated  as  °K  (degrees  Kelvin)  and  is  called  the  "color"-tempera- 
ture,  for  our  purpose,  color-temperature  may  then  be  defined  as 
the  temperature  at  which  a  complete  radiator  (black  body)  will 
match  in  color  that  of  the  source  in  question.* 

TABLE  I 

Table  Showing  the  Wavelength  of  Maximum  Energy  at  Various  Color-Temperatures 
Color  Temperature  (°K)  Wavelength  of  Maximum  Energy  (m/*) 

1000  2880 

2000  1440 

2500  1152 

3000  960 

3500  823 

4000  720 

6000  480 

8000  360 

10000  288 


*A  more  precise  definition  is:  "The  color-temperature  of  a  radiator  (source, 
lamp)  is  the  temperature  at  which  a  complete  radiator  must  be  operated  in  order 
to  emit  energy  competent  to  evoke  a  color  of  the  same  chromaticity  as  the  color 
evoked  by  the  radiant  energy  from  the  source  in  question."1 


300 


E.  M.  LOWRY  AND  K.  S.  WEAVER 


[J.  S.  M.  P.  E. 


Photographers,  whether  professional  or  amateur,  are  only  too 
well  aware  of  the  influence  which  the  quality  of  the  illumination  has 
on  the  color  rendering  of  photographic  subjects.  That  is  to  say, 
the  distribution  of  energy  from  the  source  of  light  used  is  extremely 
important  in  its  effect  upon  the  appearance  of  objects  which  it 
illuminates.  Everyone  is  familiar  with  the  phenomena  of  color 
change  brought  about  in  many  objects  when  examined  under  different 
light  sources.  The  difference  in  color-temperature  between  general- 
purpose  tungsten  lamps,  and  studio  modeling  lamps  or  between 
modeling  lamps  and  photoflood  lamps  may  well  be  the  deciding 
factor  between  correct  and  incorrect  photographic  color  reproduc- 
tion. Variations  in  the  voltage  applied  to  a  given  lamp,  or  the  length 


400 


500 


COO 


400 


500 


600 


TOO 


FIG.  2.     Diagram  showing  spectral  regions  used  in  the 
color  temperature  meter. 

of  time  it  has  been  in  operation  are  by  no  means  negligible  factors  in 
securing  and  maintaining  proper  lighting. 

In  order  that  the  photographer  may  easily  determine  the  quality 
of  lighting  which  he  is  using  and  make  suitable  adjustments  to  secure 
standard  lighting  conditions,  an  instrument  for  that  purpose,  which 
is  at  once  compact,  simple  in  operation,  and  accurate,  has  been 
developed  in  the  Kodak  Research  Laboratories. 

In  working  out  the  design  of  the  color-temperature  meter,  use  was 
made  of  some  of  the  well  known  principles  of  color  and  color  mixture. 
Suppose  that  in  some  way  we  block  out  or  absorb  certain  components 
of  the  light  from  an  incandescent  source  as  represented  by  the  shaded 
areas  in  Fig.  2  (B).  We  have  left  narrow  bands  in  the  green  and  red 
which,  if  combined,  will  produce  a  visual  impression  of  yellow  or 
orange.  This  orange  will  appear  to  the  eye  as  though  the  shaded 


Mar.,  1939]  A  COLOR-TEMPERATURE  METER  301 

areas  in  Fig.  2  (C)  had  been  removed  and  only  a  single  narrow  strip 
allowed  to  pass.  In  other  words,  a  mixture  of  a  properly  selected 
narrow  band  of  wavelengths  from  the  green  region  with  another  from 
the  red  will  produce  the  same  visual  impression  as  that  of  a  single 
small  group  in  the  yellow-orange  portion  of  the  continuous  spectrum. 

In  the  Kodak  color-temperature  meter  the  selection  of  the  proper 
portions  of  the  spectrum  is  accomplished  by  means  of  carefully  con- 
structed light  niters.  One  of  these  niters  is  so  made  up  that  it 
possesses  two  transmission  bands  with  maximum  transmittance  at 
about  520  mju  and  at  680  mju,  respectively.  The  second  filter  is  so 
composed  that  the  wavelength  of  its  maximum  transmittance  is  at 
approximately  580  mju.  The  relative  transmittances  of  the  bands  in 
the  two-band  or  dichroic  filter  are  so  adjusted  that,  when  examined 
by  light,  for  example,  that  from  a  tungsten  lamp  operating  at  a  color 
temperature  of  2100°K,  its  color  will  be  the  same  as  that  of  the  filter 
with  its  maximum  transmittance  at  580  m^u.  For  color-temperatures 
higher  than  2100°K,  the  dichroic  filter  will  appear  more  green  than 
the  monochromatic  one,  while  for  temperatures  lower  than  2100°K, 
it  will  appear  more  red.  This  property  of  dichroic  materials  was 
shown  by  Pfliiger2  in  his  work  on  anomalous  dispersion  and  was 
discussed  by  Wood  in  his  Physical  Optics* 

The  reason  for  this  behavior  may  be  explained  by  reference  to  Fig. 
3.  Curve  A  represents  the  spectrophotometric  transmission  curve 
of  the  two-band  filter  which  has  two  maxima,  at  520  m^c  and  at  680 
mju,  respectively.  Curve  B  is  that  of  the  monochromatic  filter 
which  possesses  a  transmittance  maximum  at  580  m/x.  As  stated 
above,  the  relative  transmittances  of  the  two  bands  for  filter  A  have 
been  so  adjusted  that,  when  examined  by  light  from  a  source  operat- 
ing at  a  color  temperature  of  2100°K,  this  filter  will  appear  to  be  the 
same  color  as  filter  B.  The  curve  labeled  2100°K  represents  the 
relative  energy  emitted  at  the  various  wavelengths  of  the  visible 
spectrum  by  the  source  operating  at  2100° K,  which  is  the  tempera- 
ture at  which  the  filters  will  be  color-matched.  A  source  working 
at  a  color- temperature  of  3200  °K  will  emit  energy  of  somewhat 
different  distribution  at  the  various  wavelengths.  Examination  of 
the  two  distribution  curves  shows  that  the  energy  emitted  is  relatively 
higher  at  520  m/u  and  relatively  less  at  680  m/x  for  3200°K  than  for 
the  2100°K  source.  This  will  result  in  more  light  passing  through 
the  520  rnju  band  in  A  and  less  at  680  m/x,  and  there  will  be  a  change 
in  the  color  of  the  filter  such  that  it  will  appear  more  green  than 


302 


E.  M.  LOWRY  AND  K.  S.  WEAVER 


[J.  S.  M.  P.  E. 


when  examined  with  the  2100°K  source.  In  the  case  of  filter  B, 
however,  there  will  be  relatively  little  change  in  hue  and  it  will  still 
be  yellow.  If  the  color-temperature  of  the  source  is  reduced  below 
2100°K,  for  example,  to  2000°K,  the  filter  A  will  appear  more  red 
than  B  because  the  ratio  of  energies  in  the  regions  of  the  spectrum  at 
the  positions  of  maximum  transmittance  of  the  filters  has  changed 
so  that  there  is  relatively  more  energy  in  the  red  portion  than  at  the 
original  match  point,  namely,  at  2100°K. 


TOO  eoo 

WAVt    UCNC.TH 


FIG.  3.  Spectrophotometric  transmittance 
curves  of  field  niters  used  in  color-temperature 
meter  and  relative  energy  distributions  for  color- 
temperatures  of  2100°K,  3200°K,  and  5000°K. 


In  order  that  the  two  filters  shall  remain  color-matched  when  the 
color- temperature  of  the  source  is  other  than  2100°K,  it  is  necessary 
to  modify  the  energy  distribution  of  the  source  in  some  way.  This 
modification  may  be  accomplished  by  changing  the  voltage  applied 
to  the  lamp  until  its  color-temperature  is  once  more  that  of  the 
original.  The  necessary  change  in  voltage  may  be  used  as  a  measure 
of  the  difference  from  2100°K.  Another  method  of  accomplishing 
the  desired  result  is  to  absorb  a  portion  of  the  radiant  energy  selec- 
tively with  respect  to  wavelength  in  such  a  way  that  the  remainder 
matches  that  at  the  initial  temperature.  Filters  of  this  type,  such  as 
the  so-called  daylight  glasses  or  the  Wratten  Photometric  Series  of 


Mar.,  1939] 


A  COLOR-TEMPERATURE  METER 


303 


filters,  are  well  known.  In  the  present  instance,  we  are  interested 
in  reducing  the  color-temperature,  since  the  match  point  for  the  filters 
is  lower  than  that  of  most  practical  light-sources,  and  we  require  an 
amber  colored  filter.  This  amber  filter  is  made  up  in  the  form  of  a 
wedge  and  the  amount  of  selective  absorption  is  dependent  upon  the 
thickness  of  the  wedge  at  any  point.  The  greater  the  thickness  of 
the  portion  of  the  wedge  used,  the  greater  is  the  reduction  in  color- 
temperature  of  the  light  transmitted. 

In  the  color-temperature  meter,  the  principles  just  described  have 
been  applied  as  illustrated  in  Fig.  4.  A  circular  photometric  field  P, 
with  a  fine  dividing  line  across  the  center,  is  formed  by  the  narrow 
band-filters  whose  absorption  characteristics  are  illustrated  in  Fig.  3. 


-U«HT  SOURCE: 

-PHOTOMETRIC  riC 

-CVCPIECC  I-CMS 

-WEDGE 

-KNURUED    KNOB 
-SCALE 


FIG.     4.     Illustrative 


diagram 
meter. 


of    color-temperature 


The  left  half  of  the  field,  which  is  shown  in  detail  at-B,  is  formed  by  the 
dichroic  filter,  and  the  right  half  is  formed  by  the  monochromatic  one. 
Between  the  eyepiece  lens  E  and  the  test  field  is  an  amber  wedge  W, 
for  the  purpose  of  modifying  the  energy  distribution  of  the  light 
from  the  source  being  examined.  This  wedge  is  circular  and  the 
portion  of  the  wedge  to  be  used  is  selected  by  means  of  a  small 
knurled  knob  K.  The  scale  of  the  instrument  5  is  so  calibrated  that 
it  reads  directly  the  color-temperature  of  the  source  investigated. 

Fig.  5  is  a  photograph  of  the  instrument  depicting  both  front  and 
side  views.  Comparison  of  the  reproduction  of  the  meter  with  the 
six-inch  rule  at  the  bottom  of  the  picture  illustrates  the  compactness 
and  convenient  size  of  the  design. 

Actual  operation  of  the  meter  is  accomplished  by  the  observer 
directing  the  visual  axis  of  the  instrument  (dotted  line  in  Fig.  4) 
toward  the  source  in  question.  He  then  observes  whether  the  two 


304 


E.  M.  LOWRY  AND  K.  S.  WEAVER 


[J.  S.  M.  P.  E. 


halves  of  the  field  of  view  are  color-matched  and,  if  they  are  not, 
adjusts  the  position  of  the  wedge  until  such  a  color-match  is  obtained. 
A  clockwise  motion  of  the  wedge  increases  the  amount  of  absorption 
while  a  counterclockwise  motion  decreases  it.  The  farther  the 
wedge  must  be  inserted,  the  higher  is  the  corresponding  color-tem- 
perature as  read  from  the  scale. 

Because  of  the  fact  that  there  are  certain  slight  differences  between 
the  eyes  of  different  individuals,  the  dichroic  and  monochromatic 
filters  are  not  always  color-matched  at  the  same  color  temperature. 
For  this  reason,  some  means  of  compensation  must  be  provided  if 
determinations  made  with  the  instrument  are  to  be  in  satisfactory 


FIG.  5.     Photograph  of  color-temperature  meter. 

agreement  for  two  or  more  observers.  To  overcome  this  difficulty, 
an  accommodation  scale  has  been  provided  which  enables  each  in- 
dividual to  select  the  initial  setting  of  the  amber  wedge  which  suits 
his  particular  eye.  Before  making  any  measurements,  each  ob- 
server must  set  the  scale  of  the  instrument  at  the  value  corresponding 
to  a  source  of  known  color-temperature.  A  tungsten  lamp  which 
has  been  calibrated  properly  would  serve  admirably  for  this  purpose 
but,  since  such  a  lamp  must  be  operated  at  constant  voltage,  auxiliary 
equipment  is  required  which  is  not  always  available.  Beeswax 
candles,  such  as,  for  example,  the  XX,X  Superior  Candles  made  by 
the  Socony  Vacuum  Oil  Company,  are  easily  obtained,  and,  since 
they  possess  fairly  uniform  temperature  characteristics  (1935°K  =*= 
10°),  they  are  quite  suitable  for  the  purpose  of  adjusting  the  accom- 


Mar.,  1939]  A  COLOR-TEMPERATURE  METER  305 

modation  scale,  when  used  with  the  auxiliary  blue  filter.  This 
filter,  which  raises  the  color-temperature  of  the  candle-flame  to  a 
point  above  2100°K  is  supplied  in  an  easily  attached  mount. 

To  make  the  initial  adjustment,  the  operator  first  sets  the  point 
on  the  scale  marked  C  opposite  the  index.  Then,  while  applying 
pressure  to  the  scale  with  the  thumb  of  one  hand,  to  prevent  any 
displacement  of  the  scale  relative  to  the  index,  he  rotates  the  knob 
with  the  other  hand  until  a  color-match  is  obtained  in  the  field  of 
view.  During  this  operation,  the  candle-flame  is  the  illuminant. 
After  the  preliminary  adjustment,  the  meter  is  in  condition  for  read- 
ing the  color-temperature  of  some  unknown  source. 

TABLE  II 

Table  Showing  Precision  of  Measurements  with  the  Color-Temperature  Meter 

Average  Departure  from 
Observer  Temperature  °K        Mean  of  10  Settings 

EML  2360  14 

AS  2360  24 

KSW  2360  15 

EML  2660  15 

AS  2660  30 

KSW  2660  22 

EML  2850  22 

AS  2850  26 

KSW  2850  28 

EML  3200  21 

AS  3200  35 

KSW  3200  34 


The  precision  of  the  measurements  made  with  the  color  tempera- 
ture meter  depends  upon  certain  fundamental  requirements.  In  the 
first  place,  the  operation  of  the  instrument  is  based  upon  the  ability 
of  an  observer  to  do  color-matching  and  therefore  assumes  his  color 
vision  to  be  normal ;  that  is,  he  must  not  be  color-blind  or  have  any 
noticeable  deficiencies  in  color  vision.  In  the  use  of  the  meter,  as 
in  all  operations  requiring  the  application  of  optical  instruments,  the 
precision  of  setting  is  considerably  improved  by  practice.  The  first 
few  attempts  to  balance  the  field  by  an  individual  unskilled  in  this 
type  of  measurement  are  likely  to  show  very  erratic  results,  but  as 
he  becomes  accustomed  to  the  manipulations  necessary,  his  repeat- 
ability will  improve  and  his  results  will  be  quite  satisfactory. 


306  E.  M.  LOWRY  AND  K.  S.  WEAVER 

In  Table  II  are  shown  the  average  deviations  from  the  mean  of 
ten  settings  made  by  each  of  three  observers  at  the  color  temperatures 
indicated. 

A  number  of  devices  embodying  somewhat  the  same  principles  as 
those  applied  in  the  Kodak  color  temperature  meter  have  been 
developed  over  a  period  of  years.4'5'6'7  Among  them  is  the  Harrison 
color  meter,  manufactured  by  Harrison  and  Harrison,  Hollywood, 
Calif.,  which  appeared  on  the  market  several  years  ago.  This 
instrument  is  said  to  be  useful  for  selecting  the  proper  compensating 
filter,  several  of  which  are  supplied  with  the  instrument,  to  be  used 
over  the  camera  lens  in  order  that  the  quality  of  light  reaching  the 
film  may  be  properly  adjusted  to  give  correct  color  rendering.  In 
the  Harrison  meter,  rotation  of  a  dial  causes  the  field  of  view  to  change 
from  a  blue  through  a  pink  to  a  deep  magenta.  A  setting  for  the 
selection  of  the  correct  filter  is  based  upon  the  ability  of  the  operator 
to  decide  when  the  field  "just  turns  pink." 

Another  appliance  is  that  described  in  U.  S.  Patent  1,865,878, 
issued  to  Gerhard  Naeser  and  assigned  to  the  Kaiser  Wilhelm  Insti- 
tute in  Germany.  The  Kodak  color  temperature  meter  possesses 
the  advantage  of  providing  automatic  control  of  the  brightness 
match  in  the  field  of  view,  whereas  in  the  Naeser  instrument,  the 
brightness,  if  matched  at  one  color-temperature,  will  not  match  at 
any  other.  The  Kodak  meter  has  a  further  advantage  not  provided 
by  Naeser's  arrangement.  This  is  that  the  hue  in  the  two  halves  of 
the  photometric  field  matches  at  all  temperatures  within  the  range 
of  the  instrument  while  that  of  Naeser  does  not. 

It  is  the  opinion  of  the  authors  that  the  instrument  described  in 
this  paper  provides  a  means  whereby  the  photographer  may  easily 
and  accurately  estimate  the  color- temperature  of  his  light-sources. 

REFERENCES 

1  PRIEST,  I.  G.:    J.  Opt.  Soc.  Amer.,  23  (1933),  p.  41. 

2  PFLUGER,  A.:    Ann.  der  Physik  und  Chem.,  46  (1895),  p.  412;   58  (1896),  p. 
070. 

3  WOOD,  R.  W.:    "Physical  Optics,"  p.  438  (1911). 

4  Soci6t6  Arnoux,  Vve  Chauvin  et  Cie,  French  Pat.  734,726. 

5  Siemens  &  Halske  A.-G.,  Austrian  Pat.  75,291. 
8  Siemens  &  Halske  A.-G.,  French  Pat.  715,580. 

7  Naeser,  G.,  assigned  to  Kaiser  Wilhelm  Inst.,  Germany.    U.  S.  Pat.  1,865,878. 


CHEMICAL  ANALYSIS  OF  AN  MQ  DEVELOPER* 

R.  M.  EVANS  AND  W.  T.  HANSON,  JR.** 


Summary. — The  maintenance  of  developer  activity  over  a  long  period  of  time  is 
among  the  most  important  problems  of  a  motion  picture  laboratory.  The  developer 
is  oxidized  by  the  silver  halide  in  the  emulsion  and  by  air.  When  known  amounts  of 
these  two  oxidizing  materials  react  with  the  developer,  simple  calculations  (presented 
in  a  previous  paper)  are  sufficient  to  determine  the  equilibrium  condition  of  the  de- 
veloper as  well  as  the  replenisher  formula  to  give  a  chosen  equilibrium.  Under 
ordinary  conditions  there  are  large  variations  in  the  amount  of  developer  oxidation. 
A  chemical  analysis  immediately  detects  any  deviation  from  the  correct  equilibrium 
and  permits  readjustment  of  the  replenisher  formula.  Chemical  analyses  are  pre- 
sented which  require  a  minimum  of  equipment  and  time.  In  most  cases  ease  of 
manipulation  and  speed  have  been  considered  as  more  important  factors  than  a  high 
degree  of  accuracy  but  in  all  cases  the  methods  are  capable  of  giving  results  to  an  ac- 
curacy of  five  per  cent  or  better.  Whenever  possible  the  analyses  are  colorimetric  in 
nature,  the  measurements  being  made  on  an  instrument  called  an  opacimeter.  One 
operator  can  make  a  complete  analysis  in  about  half  an  hour.  Analysis  for  any  one 
constituent  may  be  made  in  a  much  shorter  time.  It  is  emphasized  that  no  one  control 
variable  is  significant  for  specifying  the  activity  of  a  developer.  Sensitometric  curves 
are  included  demonstrating  the  time  lag  in  pH  equilibrium  but  not  in  photographic 
equilibrium  when  hydroxide  is  added  to  or  released  in  the  developer.  The  aim  of 
chemical  control  is  to  insure  a  constant  condition  of  the  developer  and  thus  constant 
photographic  quality,  rather  than  to  determine  the  degree  of  development. 

In  the  continuous  processing  of  motion  picture  film  the  developer 
can  not  be  fresh  for  each  roll  of  film  but  must  be  used  continuously. 
For  this  reason  the  developer  must  be  kept  at  constant  activity  by 
continuous  addition  of  fresh  developer  and  removal  of  old.  The 
problems  of  keeping  this  developer  at  the  correct  activity  and  of 
determing  whether  it  is  at  the  correct  activity  are  among  the  most 
important  problems  which  confront  a  motion  picture  laboratory. 

During  the  use  of  a  developer  solution,  some  of  it  is  oxidized  by  the 
film  being  developed  and  some  by  air.  If  each  of  these  reactions  is 
taking  place  at  a  constant  rate  and  the  developer  is  being  replenished 

*Presented  at  the  1938  Fall  Meeting  at  Detroit,  Mich. ;  received  October  28, 
1938.    Communication  No.  696  from  the  Kodak  Research  Laboratories. 
**Eastman  Kodak  Company,  Rochester,  N.  Y. 

307 


308  R.  M.  EVANS  AND  W.  T.  HANSON,  JR.      [j.  s.  M.  p.  E. 

at  a  constant  rate,  the  concentrations  of  the  ingredients  of  the  solu- 
tion, both  those  added  in  the  replenisher  and  those  formed  by  the 
reactions,  will  reach  an  equilibrium  concentration.  These  equilibrium 
concentrations  can  be  calculated  easily  by  means  of  equations  pre- 
sented by  the  senior  author  in  a  previous  paper. l  From  similar  cal- 
culations, it  is  possible  to  determine  the  formula  and  rate  of  addition 
of  a  replenisher  which  will  give  almost  any  desired  equilibrium  condi- 
tion. Consequently,  if  the  relative  ratio  of  silver  halide  development 
and  oxidation  by  air  can  be  kept  constant,  the  developer  solution  can 
be  brought  to  the  correct  condition  with  ease  and  will  remain  in  this 
condition.  The  resulting  photographic  quality  will  be  correct  and 
constant. 

The  problem  is  not  quite  as  simple  as  this,  however.  A  1000-ft. 
roll  of  motion  picture  positive  film  contains  about  75  to  80  grams  of 
silver  bromide  and,  depending  on  the  type  of  scenes  included  on  such 
a  roll,  5  to  60  per  cent  of  it  may  be  developed.  At  other  times  it 
may  be  necessary  to  run  leader  rather  than  exposed  film  through  the 
machine.  In  addition  to  the  variation  in  the  amount  of  silver 
bromide  which  is  developed  in  a  given  time,  there  will  be  variations 
in  the  amount  of  aeration  of  the  developer.  A  small  amount  of  air 
is  always  carried  into  the  developer  by  the  film  and  by  general  surface 
agitation,  but  this  amount  of  aeration  may  be  increased  suddenly  to 
10  to  15  times  its  original  value  by  a  leak  in  the  pumping  system. 
Even  after  a  few  days  of  continuous  processing,  these  inconsistencies 
make  it  impossible  to  know  the  concentration  of  the  important  in- 
gredients of  a  developer  solution  without  the  aid  of  chemical  analyses. 
An  analysis  of  such  a  partially  exhausted  developer  immediately 
detects  any  deviation  from  the  proper  equilibrium  condition  and  a 
new  and  correct  replenisher  formula  can  be  calculated  and  put  into 
use.  By  frequent  analysis  and  replenisher  correction,  accurate 
developer  maintenance  can  be  accomplished. 

The  number  of  analyses  necessary  for  such  developer  maintenance 
is  relatively  small  compared  with  the  total  number  of  constituents 
of  a  partially  exhausted  developer  solution  because  many  of  the 
materials  formed  during  the  oxidation  of  the  developer  are  inter- 
related and  are  formed  in  amounts  which  have  a  constant  ratio  to 
each  other.  For  example,  in  an  ordinary  MQ  developer,  the  amount 
of  bromide  formed  is  related  to  the  amount  of  elon  used  up,  the 
amount  of  sulfate  formed  is  proportional  to  the  amount  of  oxygen 
which  the  solution  has  absorbed  from  the  air,  which  in  turn  is  almost 


Mar.,  1939]          CHEMICAL  ANALYSIS  OF  MQ  DEVELOPER  309 

proportional  to  the  hydroquinone  which  is  used  up.  In  addition, 
most  of  the  oxidation  products  of  the  developer  solution  play  an 
insignificant  part  in  the  development  process  (excepting  bromide 
and  acid  or  alkali)  and  need  not  be  determined. 

In  the  case  of  an  ordinary  MQ  developer,  the  most  important 
variables  are  elon,  hydroquinone,  bromide,  sulfite,  alkaline  salts, 
and  pH.  Analyses  for  these  will  tell  the  general  condition  of  the 
bath  and  indicate  what  changes  in  the  replenisher  are  necessary. 
In  certain  special  cases,  there  may  be  other  important  ingredients 
which  should  be  determined,  but  such  cases  are  rare.  The  usual 
case  does  not  even  require  repeated  analyses  for  all  of  the  materials 
mentioned  above.  Frequent  checks  of  two  or  three  of  these  vari- 
ables, alternating  so  that  no  one  is  omitted  for  a  long  period  will  give 
sufficient  information  to  enable  correct  developer  maintenance.  In 
this  way,  the  problem  of  developer  maintenance  can  be  made  rela- 
tively simple. 

If  attempts  are  made  to  carry  the  simplification  too  far,  the  value 
of  analytical  methods  may  be  destroyed.  The  action  of  a  developer 
compound  is  affected  by  a  large  number  of  variables  which  are  by 
no  means  independent.  In  fact,  it  is  safe  to  say  that  there  is  no 
single  variable  which  may  be  used  alone  to  control  a  photographic 
developer.  The  action  of  all  developers  is  normally  a  steep  function 
of  pR  and  bromide  concentration  and  even  at  a  given  pH,  different 
developers  behave  so  differently  that  no  developer  formula  can  be 
devised  which  may  be  specified  by  a  knowledge  of  one  of  the  con- 
stituents. In  an  earlier  paper2  the  authors  showed  that  pH,  which 
has  at  times  been  considered  as  a  significant  single  variable,  may 
vary  in  the  opposite  direction  to  the  developing  properties  of  a  given 
solution.  Another  example  of  the  lack  of  correlation  between  pH. 
and  developer  action  is  shown  in  Fig.  1.  The  curves  in  (a)  were 
obtained  from  sensitometric  strips  of  motion  picture  positive  de- 
veloped in  fresh  D16  developer.  The  £H  of  this  solution  was  9.9. 
To  a  liter  of  this  solution  2  grams  of  sodium  hydroxide  were  added 
and  more  strips  were  developed  immediately  (curves  b).  The  pH 
at  this  time  was  10.1  and  the  activity  of  the  developer  had  increased 
appreciably,  as  may  be  seen  by  a  comparison  of  (a)  and  (b).  After 
this  solution  had  stood  for  7  hours,  strips  (c)  were  developed  and  the 
pH.  was  found  to  be  10.3.  A  comparison  of  (b)  and  (c)  will  show  that 
the  photographic  action  of  the  developer  had  not  changed  at  all, 
while  its  pH  had  changed  0.2  units.  Hydroxide  and  acid  are  released 


310 


R.  M.  EVANS  AND  W.  T.  HANSON,  JR.      [J.  S.  M.  P.  E. 


continuously  during  the  use  of  a  bath.  The  above  test  shows  that 
hydroxide  comes  to  an  immediate  photographic  equilibrium  with 
the  bath  but  that  this  equilibrium  does  not  have  its  full  effect  on  the 
pH  of  the  solution  for  several  hours.  Under  these  conditions  a 
measurement  of  ^>H  is  not  even  significant  as  a  measure  of  the  re- 
leased hydroxide  when  it  is  measured  during  the  use  of  the  bath. 

On  the  other  hand,  as  complete  a  chemical  analysis  of  a  developer 
as  is  possible  at  the  present  time  will  not  specify  its  exact  photo- 
graphic properties.  Traces  of  materials,  such  as  sulfide,  which  may 
be  formed  in  the  solution  by  bacteria,  will  cause  fog  even  if  present 
in  quantities  much  too  small  to  be  detected  by  any  available  analyti- 
cal methods.  Small  traces  of  certain  materials,  such  as  hypo  or 


OO     EKPOSUBE 


FIG.  1.  Sensitometric  curves  of  strips  of  motion  picture  positive  film 
developed  (a)  in  D16;  (b)  in  D16  plus  sodium  hydroxide;  (c)  same  as  (6) 
after  standing  7  hours. 

iodide,  may  accelerate  development.  On  the  other  hand,  certain 
decomposition  products  of  gelatin  may  act  as  inhibitors,  slow  down 
development,  and  decrease  fog.  Many  other  materials  may  inad- 
vertently enter  a  developer  solution  and  affect  the  development 
characteristics.  An  analysis  which  would  detect  such  traces  would 
not  only  be  extremely  tedious  but  also  unnecessary  since  the  usual 
sensitometric  control  strip  gives  that  information  quite  simply  and 
directly. 

Chemical  analyses,  then,  are  not  satisfactory  when  used  in  place 
of  photographic  tests  but  are  a  necessary  supplement  to  them.  Their 
role  is  not  to  determine  the  time  of  development  or  the  temperature 
that  is  required  to  give  a  given  contrast  but  to  insure  constant  and 
reproducible  photographic  quality.  No  matter  to  what  extent 


Mar.,  1939]          CHEMICAL  ANALYSIS  OF  MQ  DEVELOPER  311 

changes  in  the  ingredients  of  a  developer  solution  may  affect  the 
photographic  results,  the  following  statement  must  certainly  be 
true:  If  all  of  the  important  constituents  of  a  developer  solution 
are  held  constant,  the  photographic  quality  of  images  developed  in 
that  solution  will  be  constant.  Thus,  the  aim  of  the  processing 
control  should  be  to  control  the  developer  solution  itself  as  well  as  to 
control  the  degree  of  development  as  indicated  by  sensitometric  test 
strips. 

The  analytical  methods  presented  below  have  been  worked  out 
at  the  Kodak  Research  Laboratories  and  most  of  them  have  been  in 
use  over  two  years.  In  most  cases,  ease  of  manipulation  and  speed 
have  been  considered  as  more  important  factors  than  a  high  degree  of 
accuracy.  In  all  cases,  the  methods  are  capable  of  giving  results  to 
an  accuracy  of  5  per  cent  or  better.  A  500-cc.  sample  of  developer 
solution  is  sufficient  for  a  complete  analysis  and  one  operator  can 
make  a  complete  analysis  in  about  half  an  hour.  Analysis  for  any 
one  constituent  may  be  made  in  a  much  shorter  time.  The  labora- 
tory space  required  for  these  tests  is  small  and  little  equipment  is 
necessary.  Most  of  the  tests  depend  on  colorimetric  measurements 
and  may  be  run  very  efficiently  in  an  optical  instrument  called  an 
opacimeter.  In  this  instrument  a  high-aperture  lens  system  pro- 
vides a  light-beam  of  fixed  characteristics.  In  this  beam,  glass 
vessels  containing  the  products  of  the  color-forming  reactions  can  be 
placed.  The  change  in  light-intensity  is  measured  by  means  of  a 
photosensitive  cell  whose  output  is  measured  directly  by  a  microam- 
meter.  Test  reactions  can  be  run  either  in  Kohle  flasks  in  the 
instrument,  or,  when  it  is  desirable  to  use  small  amounts  of  reaction 
solutions,  in  calibrated  test  tubes.  The  opacimeter  provides  a 
sufficiently  high  level  of  illumination  so  that  narrow-band  color- 
filters  can  be  placed  in  the  light-beam  in  order  to  enhance  the  effect 
on  the  photocell  when  a  weaky  colored  reaction  mixture  is  used. 
This  instrument  is  more  fully  described  in  an  accompanying  paper. 

The  presence  of  iodide  in  a  developer  solution  has  been  mentioned 
but  no  test  for  it  is  given.  Although  it  is  known  that  small  amounts 
of  iodide  do  have  a  photographic  effect,  an  analysis  for  the  small 
amounts  present  in  an  ordinary  developer  solution  is  tedious  and  time- 
consuming.  Analyses  made  at  these  Laboratories  by  Ballard  and 
Yutzy  have  shown  that  the  equilibrium  amount  of  iodide  in  a  developer 
which  is  used  for  positive  film  is  only  about  0.0005  gram  per  liter  (ex- 
pressed as  potassium  iodide)  and  for  negative  film  is  only  0.0015  to 


312  R.  M.  EVANS  AND  W.  T.  HANSON,  JR.      [j.  s.  M.  P.  E. 

0.0020  gram  per  liter.  In  both  cases,  equilibrium  is  reached  when  less 
than  5  feet  of  film  per  liter  of  solution  has  been  developed.  In  the 
usual  case,  the  equilibrium  is  rapidly  established  and  the  total  amount 
of  iodide  present  does  not  vary  appreciably.  There  is  no  real  necessity 
for  analysis.  It  is  recommended,  however,  that  when  entirely  fresh 
solutions  are  placed  in  a  machine,  an  amount  of  potassium  iodide 
approximately  equivalent  to  the  above  figures  should  be  added  to  the 
solution  and  a  small  amount  of  fogged  film  (perhaps  one  foot  per 
gallon)  should  be  run  through  the  solution. 

Elon. — The  analysis  for  elon  is  based  on  the  formation  of  a  yellow 
solution  when  elon  reacts  with  furyl  acrolein.  Since  elon  sulfonate 
which  is  present  as  an  oxidation  product  undergoes  the  same  reac- 
tion, the  elon  must  be  separated  from  the  developer  solution  by  ex- 
traction. Lehmann  and  Tausch3  have  published  an  analytical 
method  for  elon  in  which  the  elon  was  extracted  for  several  hours  in  a 
continuous  extraction  apparatus  and  then  determined  by  iodo- 
metric  titration.  This  method  gives  accurate  results  but  is  too 
time-consuming  to  be  used  as  a  control  method.  While  the  method 
outlined  here  is  not  as  accurate  as  the  Lehmann  and  Tausch  method, 
it  has  the  merit  that  it  is  rapid  and  it  is  felt  that  in  many  cases  rapid- 
ity is  more  important  than  extreme  accuracy.  One  objection  to  the 
present  method  is  that  the  results  are  affected  by  small  deviations 
from  the  exact  procedure  and  the  analyst  must  become  thoroughly 
familiar  with  the  details  of  the  method  before  satisfactory  results 
can  be  obtained.  However,  satisfactory  results  can  be  obtained  and 
in  the  absence  of  a  more  easily  controlled  analysis  the  present  method 
is  presented. 

STOCK  SOLUTIONS  AND  CHEMICALS 

(1)  Buffer  at  pU  8.4 
Trisodium  phosphate  150  gm. 
Water  960  cc. 
Cone,  hydrochloric  acid  40  cc. 

(2)  Furyl  Acrolein  10  gm. 
Ethyl  ether  200  cc. 
Filter 

(This  solution  is  stable  only  a  few  weeks  and  should  not  be  made  up  in  large 

amounts  nor  should  the  solid  be  stored  for  long  periods  of  time.) 

Ethyl  acetate 

Concentrated  hydrochloric  acid 

Solid  sodium  carbonate 

Phenolphthalein 

Solid  sodium  chloride 


Mar.,  1939]          CHEMICAL  ANALYSIS  OF  MQ  DEVELOPER  313 

EQUIPMENT 

One  250-cc.  separatory  funnel 

One  250-cc.  Erlenmeyer  flask 

One  125-cc.  Erlenmeyer  flask 

Two  test  tubes 

Pipettes  1,  2,  5,  and  10  cc. 

One  50-cc.  graduated  cylinder 

One  mechanical  shaker 

Two  storage  bottles 

One  No.  47  Wratten  filter 

Opacimeter  or  its  equivalent 

Procedure. — To  50  cc.  of  developer  solution  in  a  250-cc.  Erlen- 
meyer flask  are  added  a  few  drops  of  phenolphthalein  and  sufficient 
concentrated  hydrochloric  acid  to  destroy  the  pink  color.  An 
addition  of  solid  sodium  carbonate  to  just  restore  this  color  is  made 
and  then  50  cc.  of  the  stock  buffer  solution  No.  1,  30  gm.  of  solid 
sodium  chloride,  and  50  cc.  of  ethyl  acetate  are  added.  This  is 
sufficient  sodium  chloride  to  form  a  saturated  solution.  The  mix- 
ture is  shaken  for  five  minutes  on  a  mechanical  shaker.  The  pink 
color  of  the  phenolphthalein  disappears  during  this  shaking.  The 
mixture  is  poured  into  a  separatory  funnel  and,  after  a  few  minutes, 
the  water  layer  is  drawn  off.  From  2  to  10  cc.  of  the  ethyl  acetate 
solution  are  added  to  25  cc.  of  water  and  2  cc.  of  concentrated  hydro- 
chloric acid  in  a  125-cc.  flask  to  which  is  also  added  10  cc.  of  the  furyl 
acrolein  stock  solution  No.  2.  This  is  shaken  for  five  minutes  on  the 
mechanical  shaker  and  then  allowed  to  stand  for  five  minutes  in  a 
separatory  funnel.  The  water  layer  at  the  bottom  is  poured  into  a 
standard  test  tube  and  its  transmission  is  read  on  the  opacimeter 
through  a  No.  47  Wratten  filter.  The  elon  concentration  is  de- 
termined from  an  empirical  calibration  curve  made  up  by  analyzing 
known  fresh  solutions. 

The  total  amount  of  elon  plus  elon  sulfonate  which  is  present  can 
be  determined  by  allowing  1  to  5  cc.  of  the  original  developer  solution 
(in  place  of  the  ethyl  acetate  extract)  to  react  with  the  furyl  acrolein. 
A  calibration  curve  must  be  prepared  following  this  procedure. 

Because  of  the  sensitivity  of  this  test  to  small  changes  in  the  con- 
ditions under  which  it  is  carried  out,  each  step  in  the  procedure  must 
be  carefully  controlled.  The  extraction  of  elon  is  a  function  of  both 
pH  and  total  salt  content,  so  both  of  these  factors  must  be  reproduced 
accurately.  The  pH  of  the  stock  buffer  solution  must  not  vary  more 


314  R.  M.  EVANS  AND  W.  T.  HANSON,  JR.      [j.  s.  M.  P.  E. 

than  0.1  from  8.4  and  the  amount  of  sodium  chloride  added  before 
the  extraction  must  be  sufficient  to  saturate  the  solution.  The 
reaction  between  elon  and  furyl  acrolein  precedes  slowly  and  suffi- 
cient time  for  its  completion  can  not  be  allowed,  so  the  time  of  shak- 
ing at  this  stage  of  the  analysis  as  well  as  the  time  of  standing  in  the 
separatory  funnel  after  shaking  must  be  carefully  reproduced.  With 
these  precautions  properly  followed  and  a  calibration  curve  pre- 
pared for  a  given  developer  formula,  an  accuracy  of  about  five  per 
cent  can  be  achieved.  If  the  analysis  must  be  made  on  an  unknown 
developer  formula,  less  accurate  results  can  be  expected,  but,  if 
necessary,  the  first  rough  analysis  can  be  followed  by  a  calibration 
curve  and  another  analysis. 

Such  a  procedure  takes  about  two  or  three  hours  and  should  be 
necessary  only  in  rare  cases.  Under  ordinary  conditions  the  test 
takes  about  15  minutes. 

Hydroquinone  and  Hydroquinone  Sulfonate. — The  test  for  these 
compounds  is  based  on  the  measurement  of  the  intensity  of  the  color 
formed  when  they  are  oxidized  by  potassium  ferricyanide  in  the  pres- 
ence of  sodium  sulfite.  Both  hydroquinone  and  hydroquinone  sul- 
fonate  give  the  same  color  but  can  be  separated  by  acidifying  the 
developer  solution  and  extracting  the  hydroquinone  with  ethyl 
acetate.  The  sulfonate  remains  in  the  water  layer.  From  the 
stoichiometry  of  the  titration  with  ferricyanide  it  appears  that  the 
colored  product  is  a  semiquinone  of  hydroquinone  disulfonate  but 
no  definite  proof  of  this  has  been  undertaken.  The  exact  procedure 
is  as  follows : 

STOCK  SOLUTIONS  AND  CHEMICALS 

(1)  Sodium  sulfite  30  gm. 
Sodium  carbonate           20  gm. 
Water  to  1  liter 

(2)  Potassium  ferricyanide  0.2  molal  (approximately) 
(5)     Bromophenol  blue     (0.1  per  cent  water  solution) 

Concentrated  hydrochloric  acid 

Ethyl  acetate 

All  solutions  are  fairly  stable  and  can  be  kept  for  several  weeks.     None  of 

them  need  be  standardized. 

EQUIPMENT 

One  250-cc.  separatory  funnel 
One  50-cc.  graduate 


Mar.,  1939]          CHEMICAL  ANALYSIS  OF  MQ  DEVELOPER  315 

Pipettes  1,  2,  and  5  cc. 

One  300-cc.  Kohle  flask  with  side  arm  for  air  agitation 

One  No.  44  Wratten  filter 

Two  storage  bottles 

Compressed  air  (low  pressure) 

Procedure. — Hydroquinone.  To  50  cc.  of  developer  solution  are 
added  a  few  drops  of  bromphenol  blue  and  then  concentrated  hydro- 
chloric acid  is  added  until  the  solution  just  turns  yellow.  Fifty  cc. 
of  ethyl  acetate  are  added  and  the  mixture  shaken  by  hand  for  one 
minute  in  a  separatory  funnel.  This  is  then  allowed  to  stand  until 
two  layers  have  separated.  One  to  5  cc.  of  the  ethyl  acetate  layer 
(the  upper  layer)  are  added  to  275  cc.  of  water  and  25  cc.  of  stock 
solution  No.  2  in  the  Kohle  flask.  This  solution  is  placed  in  the 
opacimeter  and  agitated  by  air  introduced  through  a  glass  tube 
inserted  in  the  flask.  The  No.  44  filter  is  placed  in  the  beam  of  light, 
the  microammeter  set  so  that  the  reading  is  at  a  maximum,  and  ferri- 
cyanide  solution  is  poured  in  slowly  and  uniformly  until  the  deflec- 
tion of  the  galvanometer  needle  is  a  minimum.  This  point  of  mini- 
mum deflection,  when  read  on  the  calibration  curve,  gives  the  amount 
of  hydroquinone  present. 

If,  instead  of  using  1  to  5  cc.  of  the  ethyl  acetate  layer  in  the  above 
procedure,  an  equal  amount  of  the  water  layer  is  used,  the  hydro- 
quinone sulfonate  rather  than  the  hydroquinone  causes  the  color, 
and  from  the  proper  calibration  curve  its  concentration  can  be  de- 
termined. The  total  amount  of  hydroquinone  and  hydroquinone 
sulfonate  can  be  checked  by  running  the  test  on  the  original  developer 
solution  without  extraction. 

In  order  to  achieve  the  greatest  accuracy,  calibration  curves 
should  be  obtained  for  each  developer  formula  because,  although  the 
test  is  specified  for  hydroquinone  in  an  ordinary  MQ  developer, 
different  total  salt  concentrations  affect  the  extraction  and,  con- 
sequently, the  determination  of  free  hydroquinone.  Under  the, 
proper  conditions,  the  test  gives  analyses  with  an  error  of  much  less 
than  5  per  cent. 

Sodium  Sulfite. — The  test  for  sulfite  is  based  on  the  fact  that 
certain  dyes  are  quantitatively  bleached  by  sulfite  solutions. 

STOCK  SOLUTIONS 

(1)     Isopropyl  alcohol          100  cc. 
Water  900  cc. 


316  R.  M.  EVANS  AND  W.  T.  HANSON,  JR.      [J.  S.  M.  P.  E. 

Glacial  acetic  acid  1  cc. 

Acid  green  L  (No.  764)        1  gm. 

Filter  through  No.  42  filter  paper 

pH  =  3.9 
(2)  Sodium  carbonate  30  gm. 

Sodium  bicarbonate          30  cc. 

Water  to  make  1  liter 

pH  =  9.4 
Both  solutions  are  stable  for  several  months. 


EQUIPMENT 


Two  storage  bottles 

One  small  filter  funnel  and  paper 

Pipettes,  1  and  10  cc. 

Two  test  tubes 

One  No.  23  Wratten  filter 


Procedure. — To  1  cc.  of  the  developer  solution  are  added  10  cc.  of 
stock  solution  No.  2  and  10  cc.  of  stock  solution  No.  1.  This  order 
of  addition  of  these  solutions  must  be  followed  because  the  bleaching 
of  the  dye  is  affected  by  />H  and  must  be  carried  out  in  a  well  buffered 
solution.  Since  the  dye  is  not  stable  for  long  periods  of  time  if 
dissolved  in  the  buffer  itself,  it  must  be  kept  in  the  acid  solution  as 
recommended.  The  transmission  of  this  partially  bleached  dye 
solution  is  measured  on  the  opacimeter  through  a  Wratten  No.  22 
or  25  filter  and  the  sulfite  concentration  is  determined  from  a  cali- 
bration curve. 

Bromide. — Bromide  is  determined  by  titrating  with  a  standardized 
silver  nitrate  solution,  using  metanil  yellow  as  an  adsorption  in- 
dicator.* Chloride  or  any  other  material  which  forms  a  salt  less 
soluble  than  silver  chloride  will  be  included  by  this  test  and  in  the 
presence  of  unknown  amounts  of  such  materials  the  method  can  not 
be  used.  A  method  which  eliminates  these  objections  has  been 
worked  out  by  Ballard  of  these  Laboratories  and  has  been  used 
quite  successfully,  but  since  it  is  more  time-consuming  than  the 
present  method,  it  will  not  be  included  here.  Iodide  may  be  con- 
sidered constant,  as  was  pointed  out  above. 


*This  method  was  worked  out  at  the  Kodak  West  Coast  Laboratories  (Holly- 
wood, Calif.)  by  Atkinson.  Shaner,  and  Huse. 


Mar.,  1939]          CHEMICAL  ANALYSIS  OF  MQ  DEVELOPER  317 

STOCK  SOLUTION  AND  CHEMICALS 

(1)  Silver  nitrate  (standardized)  about  0.03  N 

(2)  Metanil  yellow  (0.1  per  cent  water  solution) 
Concentrated  sulfuric  acid 

EQUIPMENT 

One  50-cc.  burette  and  holder 

One  15-cc.  pipette 

One  250-cc.  Erlenmeyer  flask 

Procedure. — To  15  cc.  of  developer  solution  are  added  5  to  10  cc. 
of  concentrated  sulfuric  acid.  This  is  then  diluted  to  about  100  cc. 
and  cooled.  Two  or  three  drops  of  metanil  yellow  are  added  and 
then  titrated  with  standard  silver  nitrate  solution  until  the  solu- 
tion changes  from  purple  to  red.  The  usual  volumetric  calculations 
can  be  used  to  determine  the  amount  of  halide  present  or  a  calibra- 
tion curve  can  be  prepared.  The  change  of  color  at  the  end  point  is 
sometimes  rather  difficult  to  distinguish  but  with  a  little  experience 
under  the  proper  lighting  conditions  the  titration  can  be  controlled 
to  an  accuracy  of  two  or  three  per  cent. 

Carbonate. — The  analysis  for  carbonate  is  based  on  the  measure- 
ment of  the  pressure  developed  in  a  nearly  constant  volume  system 
held  at  constant  temperature,  when  carbon  dioxide  is  released  from 
the  developer  by  the  addition  of  a  strong  acid.  In  order  to  avoid 
the  formation  of  sulfur  dioxide  from  the  sulfite  in  the  solution, 
quinone  must  be  added.  This  reacts  with  the  sulfite  to  form  hydro  - 
quinone  monosulfonate. 


CHEMICALS 


Concentrated  hydrochloric  acid 
Solid  quinone 


EQUIPMENT 

One  500-cc.  Erlenmeyer  flask  (calibrated) 
One  stopper  bucket  (as  shown  in  Fig.  2) 
One  open-end  manometer  (filled  with  CCL.) 
One  burette 
Pipettes,  5  and  10  cc. 
Constant-temperature  water  bath 
Rubber  tubing 


318 


R.  M.  EVANS  AND  W.  T.  HANSON,  JR.      [j.  s.  M.  P.  E. 


Procedure. — Five  cc.  of  developer  solution  and  10  cc.  of  water  are 
pipetted  into  the  calibrated  Erlenmeyer  flask  and  about  0.5  gram  of 
solid  quinone  is  added.  Shake  thoroughly.  Two  cc.  of  concen- 
trated hydrochloric  acid  are  run  into  the  stopper  bucket  from  a 
burette  and  the  stopper  is  inserted  in  the  flask.  This  is  then  im- 
mersed in  a  constant  temperature  bath  and  allowed  to  come  to 

temperature  equilibrium.  The 
manometer  is  then  connected  to 
the  arm  of  the  stopper  bucket  and 
the  flask  is  tilted  so  that  the  acid 
is  poured  into  the  developer  solu- 
tion. After  it  has  been  shaken  thor- 
oughly, the  flask  is  again  brought 
to  temperature  equilibrium  and  the 
pressure  read  from  the  manometer. 
The  carbonate  concentration  is  de- 
termined from  a  calibration  curve. 

The  exact  size  of  the  flask  used  in 
the  analysis  and  the  temperature  at 
which  it  is  carried  out  are  imma- 
terial as  long  as  the  calibration 
curve  is  prepared  under  the  same 
conditions.  For  this  reason,  it 
might  be  desirable  to  prepare  cali- 
bration curves  at  several  tempera- 
tures and  then  to  use  an  ordinary 
sink  with  hot  and  cold  water  mixed 
to  give  a  fairly  constant  tempera- 
ture, as  the  constant  temperature 
bath.  Under  such  conditions,  where 
the  temperature  might  vary  0.1° 
the  test  is  accurate  to  better  than  10  per  cent.  In  most  cases  this 
accuracy  is  quite  sufficient,  especially  if  the  pH  of  the  solution  is 
determined.  However,  an  accuracy  of  3  to  4  per  cent  can  be 
obtained  by  carefully  controlling  the  condition  of  the  test. 

Sulfate. — The  sulfate  analysis  is  based  on  the  precipitation  of 
barium  sulfate  by  the  addition  of  barium  chloride  after  removal  of 
the  sulfite  and  carbonate  from  the  developer  solution.  The  amount 
of  barium  sulfate  formed  is  determined  turbidimetrically  by  means  of 
the  opacimeter. 


FIG.  2.  Stopper  bucket  and 
flask  for  use  in  carbonate  analy- 
sis. 


Mar.,  1939]          CHEMICAL  ANALYSIS  OF  MQ  DEVELOPER  319 

STOCK  SOLUTION  AND  CHEMICALS 

(1)     Barium  chloride  dihydrate  10  gm. 
Water  1  liter 

Concentrated  hydrochloric  acid 

EQUIPMENT 

One  50-cc.  graduate 

One  50-cc.  burette 

Two  test  tubes 

Pipettes  2,  10,  and  25  cc. 

One  125-cc.  Erlenmeyer  flask 

Low-pressure  steam  (if  available)  or  hot  plate 

One  filter  funnel  and  filter  paper 

Procedure. — To  25  cc.  of  developer  solution  in  an  Erlenmeyer  flask 
10  cc.  of  concentrated  hydrochloric  acid  are  added  and  the  solution 
is  boiled  or  bubbled  with  steam  for  two  or  three  minutes  in  order  to 
remove  sulfur  dioxide  and  carbon  dioxide.  This  is  cooled  and 
diluted  to  100  cc.  in  the  graduated  cylinder  and,  if  the  solution  is 
turbid  at  this  point,  it  must  be  filtered.  Two  cc.  of  this  solution  are 
pipetted  into  25  cc.  of  stock  solution  No.  1  in  a  standard  test  tube, 
a  cork  stopper  is  inserted,  and  the  tube  is  inverted,  three  or  four  times. 
Its  transmission  is  then  read  on  the  opacimeter  and  the  sulfate  con- 
centration determined  from  a  calibration  curve. 

pR. — The  pR  of  developer  solutions  can  sometimes  be  measured 
colorimetrically  by  means  of  indicators  but,  in  many  cases,  the  solu- 
tions are  colored  to  such  an  extent  that  colorimetric  observations  are 
impracticable  and  electrometric  methods  must  therefore  be  used. 
The  usual  types  of  glass  electrodes  used  in  conjunction  with  a  stand- 
ard reference  electrode  and  an  ordinary  potentiometer  are  quite 
satisfactory.  At  the  present  time,  most  glass  electrodes  have  a 
fairly  large  sodium-ion  error  at  high  pR  values  but  there  is  now  enough 
information  in  the  chemical  literature  so  that  corrections  for  such 
errors  can  be  made  with  sufficient  accuracy.  In  any  case  where 
variations  of  pR  rather  than  absolute  values  are  desired,  the  errors 
may  be  neglected.  The  most  important  use  of  pR  in  developer 
analysis  is  to  obtain  a  quick  check  in  cases  where  large  errors  of 
mixing  are  suspected  or  where  it  is  desired  to  distinguish  between  too 
greatly  different  formulas  such  as  those  used  for  a  negative  and  a 
positive  film. 


320  R.  M.  EVANS  AND  W.  T.  HANSON,  JR. 

REFERENCES 

1  EVANS,   R.   M.:    "Maintenance  of  a  Developer  by  Continuous  Replenish- 
ment," /.  Soc.  Mot.  Pict.  Eng.,  XXXI  (Sept.,  1938),  p.  273. 

2  EVANS,  R.  M.,  AND  HANSON,  W.  T.,  JR.  :    "Reduction  Potential  and  the  Com- 
position of  an  MQ  Developer,"  /.  Soc.  Mot.  Pict.  Eng.,  XXX  (May,  1938),  p. 
.559. 

3  LEHMANN,  E.,  AND  TAUSCH,  E.:    "Zum  Chemismus  der  Metol-Hydrochinon 
Entwicklung,"  Phot.  Korr.,  71  (Feb.,  1935),  p.  17;  71  (March,  1935),  p.  35. 

DISCUSSION 

MR.  DEPUE:  What  is  the  best  length  of  time  for  processing  positive  film  in 
the  machine,  two  minutes  or  three  and  a  half? 

MR.  EVANS:  That  depends  a  good  deal  upon  the  machine  and  the  precision 
with  which  you  wish  to  repeat  results.  At  two  minutes,  of  course,  it  is  much  more 
difficult  to  repeat  results  than  at  five  or  six.  I  do  not  think  it  can  be  stated 
whether  one  is  to  be  preferred  to  the  other. 

MR.  SCHMIDT:  I  understand  that  upon  the  addition  of  alkali  to  the  developer 
a  certain  time  is  required  for  the  pH  to  reach  the  definite  value  that  corresponds  to 
the  equilibrium. 

MR.  EVANS:  Yes.  In  some  cases  half  a  day  is  required  before  the  pH  equi- 
librium is  established  so  it  can  be  read. 

MR.  SCHMIDT:  Do  you  propose  an  explanation  of  that?  I  expected  that  the 
reaction  between  the  ions  would  take  place  in  a  shorter  time. 

MR.  EVANS:  We  expected  that  also,  but  it  does  not  appear  to  be.  Perhaps  it 
has  to  do  with  the  organic  constituents  that  are  present. 

MR.  CRABTREE:     Do  you  estimate  the  sulfonates  formed  in  the  developer? 

MR.  EVANS:  The  sulfonates  are  included  in  the  analysis.  They  are  not 
strictly  necessary  unless  it  is  desired  to  know  what  the  original  formula  was.  The 
sum  of  acting  developing  agents  and  their  sulfonates  is  equal  to  the  concentration 
in  the  replenisher  of  the  original  compound,  and  accordingly  it  is  sometimes  inter- 
esting to  know  whether  a  change  in  the  bath  is  due  to  the  concentration  in  the 
replenisher. 


AN  OPACIMETER  USED  IN  CHEMICAL  ANALYSIS1 
R.  M.  EVANS  AND  G.  P.  SILBERSTEIN** 


Summary. — The  opacimeter  is  an  optical  instrument  designed  to  measure  the  light- 
transmission  of  a  colored  or  turbid  solution.  A  Loewenthal  photronic  type  light- 
sensitive  cell  connected  to  a  microammeter  is  used  to  measure  the  intensity  of  the  light 
transmitted  by  the  solution  under  test.  The  light-intensity  falling  on  the  sensitive  cell 
is  kept  within  a  fixed  range  by  varying  the  distance  of  the  cell  from  the  source.  The 
instrument  is  arranged  so  that  a  30-cc.  test  tube  or  a  300-cc.  Kohle  flask  may  contain 
the  reaction  mixture.  The  results  of  analyses  are  determined  from  calibration  curves 
prepared  from  known  solutions. 

A  great  many  quantitative  tests  for  the  chemical  constituents  of 
solutions  lend  themselves  readily  to  measurements  based  on  the 
change  in  light-transmission  of  the  solution  after  addition  of  suitable 
reagents.  A  wide  variety  of  optical  instruments  designed  to  make 
measurements  of  this  sort  have  been  in  use  for  a  long  time.  The 
purpose  of  this  article  is  to  describe  a  new  design  of  such  an  instru- 
ment which  possesses  several  advantages. 

The  name  "opacimeter"  seems  a  suitable  one  to  give  to  the  instru- 
ment shown  in  Fig.  1.  A  Loewenthal  photronic  type  light-sensitive 
cell  (chosen  because  of  its  stability  and  high  sensitivity)  is  used  to 
measure  the  intensity  of  the  light  transmitted  by  the  solutions  under 
test.  This  cell  is  color-selective,  and  its  sensitivity  depends  upon  the 
intensity  level  of  the  incident  light.  For  these  reasons,  the  optical, 
system  shown  diagrammatically  is  designed  to  work  at  a  constant 
color-temperature  of  the  light-source.  The  light-intensity  falling 
on  the  Loewenthal  cell  is  kept  within  a  fixed  range  by  varying  the 
distance  of  the  cell  from  an  opal  glass  diffusing  disk  illuminated  by 
the  light  passing  through  the  solution  under  test.  In  order  to  cut  out 
the  infrared  sensitivity  of  the  light-receiver  and  also  to  prevent  heat- 
rays  from  reaching  the  test  solutions,  a  flask  containing  a  4-per  cent 
copper  sulfate  solution  is  placed  on  the  light-source  side  of  the  flask 

*  Presented  at  the  1938  Fall  Meeting  at  Detroit,  Mich.;  received  October  28, 
1938.     Communication  No.  697  from  the  Kodak  Research  Laboratories. 
**  Eastman  Kodak  Co.,  Rochester,  N.  Y. 

321 


322 


R.  M.  EVANS  AND  G.  P.  SILBERSTEIN      [j.  s.  M.  P.  E. 


used  in  making  the  tests.  A  high-aperture  lens  system  maintains  a 
parallel  beam  of  light  passing  through  the  solution  in  the  test  flask, 
thus  permitting  a  more  efficient  system.  As  long  as  the  liquid  level 
is  a  small  distance  above  the  top  of  the  light-path,  changes  in  this 
level  have  no  effect  on  the  amount  of  light  reaching  the  light-receiver. 


FIG.  1.     Diagram  of  optical  system. 

FIG.  2.     Diagram  showing  scheme  for  air  agitation  within  Kohle  flask. 
FIG.  3.     Test-tube  holder. 

The  opacimeter  is  designed  to  use  Kohle  culture  flasks  as  test  cells, 
these  being  readily  available,  easily  cleaned,  and  quite  uniform  in 
dimensions.  The  procedure  followed  is  to  make  the  calibration  run 
on  any  given  test  in  the  same  flask  as  that  to  be  used  in  testing  the 
solution.  In  case  of  breakage,  a  new  flask  may  be  calibrated  easily. 
In  Fig.  2  is  shown  an  air-stirring  system  built  into  one  of  the  Kohle 


Mar.,  1939]  OPACIMETER  FOR  CHEMICAL  ANALYSIS  323 

flasks.  This  consists  of  a  bent  glass  tube  ending  in  a  fine  orifice. 
The  large  end  of  the  tube  is  attached  to  a  source  of  compressed  air, 
the  fine  air-jet  producing  very  efficient  circulation  in  the  solution  in 
the  flask,  air  bubbles  passing  around  the  outside  of  the  flask  and  not 
entering  that  portion  of  the  solution  lying  in  the  light-path.  As  can 
be  seen  from  the  photograph  (Fig.  4)  the  test  flasks  do  not  need  to  be 
removed  from  the  instrument  in  order  to  introduce  test  liquids ;  thus, 
continuously  stirred  titrations  can  be  run  by  using  a  burette  held  in 
place  above  the  opening  of  the  air-stirring  flask  just  mentioned. 
When  it  is  desirable  to  use  a  small  amount  of  solution  in  making  a 
test,  the  Kohle  flask  can  be  replaced  by  a  test-tube  holder,  such  as 


FIG.  4.     Photograph  of  instrument,  with  light  guards  removed. 

shown  in  Fig.  3.  The  test  tube  when  filled  with  solution  forms  a 
cylindrical  lens  which,  in  conjunction  with  the  main  optics,  covers  the 
opal-glass  diffusing  disk  with  light.  Ordinary  test  tubes  can  be  used 
for  this  purpose,  provided  each  one  is  calibrated  for  the  particular  test 
with  which  it  is  to  be  used. 

In  most  cases,  tests  are  run  by  adjusting  the  position  of  the  light- 
receiver  so  that  the  microammeter  in  series  with  it  gives  a  deflection' 
of  200/z  with  a  blank  solution  in  the  test  flask  and  with  a  suitable 
Wratten  light-filter  in  the  holder  shown  in  Fig.  1.  The  decrease  in 
transmission  of  the  solution  after  the  test  reaction  has  taken  place  is 
read  on  the  meter  scale  in  microamperes.  This  reading  can  then  be 
converted  into  the  amount  of  the  particular  ingredient  by  the  use  of  a 
calibration  curve.  If  it  is  desired  to  measure  the  optical  density  of 


324  R.  M.  EVANS  AND  G.  P.  SILBERSTEIN 

the  solution,  meter  readings  can  be  converted  to  densities  from  the 
relation : 

200 
Density  =  log 


reading 

A  200-watt,  110- volt  projection  lamp  (with  a  prefocus  base  to 
minimize  adjustment  when  replacing  a  burned  out  lamp)  gives  suffi- 
cient light  for  most  tests,  even  when  such  a  low-transmission  light 
filter  as  Wratten  No.  53  is  used.  The  meter  used  in  conjunction  with 
the  Loewenthal  cell  is  a  Weston  model  No.  301  DC  microammeter, 
reading  n  at  full  scale. 

The  use  of  the  instrument  is  not  restricted  to  chemical  reactions 
giving  rise  to  a  clear,  colored  solution.  In  practice,  it  has  been  found 
that  the  amount  of  constituent  present  can  be  reliably  determined  by 
measuring  the  change  in  light- transmission  arising  from  the  formation 
of  uniformly  dispersed  precipitates,  especially  if  settling  of  the  pre- 
cipitates is  prevented  by  use  of  suitable  dispersing  agents. 

In  making  measurements  on  nonscattering  colored  solutions,  the 
use  of  colored  filters  in  the  light-beam  greatly  increases  the  sensitivity 
of  the  particular  test  being  made.  Common  procedure  in  these  cases 
is  to  choose  a  filter  having  a  strong  absorption  band  in  the  spectral 
region  where  the  transmission  of  the  colored  solution  being  measured 
is  a  maximum.  In  the  case  of  reactions  giving  several  transmission 
bands  owing  to  the  presence  in  solution  of  constituents  other  than  the 
one  under  test,  the  use  of  a  proper  color-filter  makes  it  possible  to 
separate  the  useful  transmission  band  from  the  others. 


NEW  MOTION  PICTURE  APPARATUS 


During  the  Conventions  of  the  Society,  symposiums  on  new  motion  picture  appara- 
tus are  held  in  which  various  manufacturers  of  equipment  describe  and  demonstrate 
their  new  products  and  developments.  Some  of  this  equipment  is  described  in  the 
following  pages;  the  remainder  will  be  published  in  subsequent  issues  of  the  Journal. 


A  NEW  PROJECTOR  MECHANISM* 
H.  GRIFFIN** 

During  the  past  few  years  the  Research  and  Engineering  Departments  of  this 
company  have  been  searching  for  a  device  or  devices  that  would  permit  changes 
in  the  design  of  projection  equipment  that  would  result  in  greatly  improved  film 
presentation.  Many  ideas  were  considered  and  rejected,  some  because  of  extreme 
complications,  and  all  because  nothing  was  found  that  would  compare  favorably 
so  far  as  screen  results  were  concerned,  with  equipment  of  our  earlier  designs. 
After  making  this  decision  approximately  two  years  ago,  it  was  decided  to  im- 
prove, so  far  as  we  possibly  could,  upon  what  were  our  earlier  conceptions  of  the 
best  available  mechanisms,  with  the  result  that  about  two  months  ago  there  was 
released  to  the  industry  the  company's  latest  development  in  motion  picture 
projectors,  the  Simplex  E-7  mechanism  (Fig.  1). 

In  considering  the  design  of  modern  motion  picture  projection  equipment  it  is 
necessary  to  deal  with  two  major  requirements:  (1)  greater  screen  illumination, 
without  increasing  the  light  coming  from  the  arc  or  other  illuminating  means, 
and  (2)  increased  steadiness  of  the  projected  picture.  The  means  whereby  these 
have  been  accomplished,  together  with  a  brief  summary  of  other  changes  and  im- 
provements, constitute  the  subject  of  this  paper. 

Tn  the  conventional  motion  picture  projector  mechanism  there  are  a  revolving 
shutter  either  in  the  front  or  in  the  rear  which  cuts  off  the  light  as  the  film  is 
pulled  down  past  the  aperture  plate  by  the  intermittent  movement,  and  openings 
in  the  shutter,  either  front  or  rear,  which  allow  the  light-beam  to  pass  through  the 
film  to  the  screen  while  the  film  is  at  rest  in  front  of  the  aperture  plate.  Increased 
illumination  obtained  from  the  E-7  projector  has  been  brought  about  by  placing  a 
second  shutter  in  front  of  the  lens  (Fig.  2),  this  shutter  being  attached  to  the  same 
shutter  shaft  and  revolving  in  the  same  direction  as  the  rear  shutter  or  the  shutter 
between  the  illuminant  and  the  aperture  plate.  By  this  design  it  is  possible  to 
cut  off  half  of  the  light-beam  behind  the  aperture  plate,  and  half  of  the  picture  at 

*  Presented  at  the  1938  Spring  Meeting  at  Washington,  D.  C. ;  received  April 
14,  1938. 

**  International  Projector  Corp.,  New  York,  N.  Y. 

325 


326 


NEW  MOTION  PICTURE  APPARATUS        [J.  S.  M.  P.  E. 


the  same  instant  in  front  of  the  lens,  since  the  image  is  inverted  after  passing 
through  the  lens.  We  are  thus  able  (since  it  is  not  then  necessary,  as  in  the  case 
of  one  shutter,  to  cut  the  entire  picture  from  the  screen  before  moving  the  next 
one  into  position)  to  cut  the  top  and  bottom  halves  of  the  picture  at  the  same 
instant  and  eliminate  approximately  20  degrees  from  each  shutter  blade,  and 
hence  pass  that  much  more  light  to  the  screen.  This  results  in  an  approximate  in- 
crease of  from  12  to  15  per  cent  in  screen  illumination. 


FIG.  1.     Simplex  E-7  projector. 

This  result  is  obviously  attained  without  increasing  the  speed  of  any  part  of 
the  mechanism  since,  as  before  stated,  both  shutters  revolve  at  the  same  speed 
(1440  rpm.),  being  attached  to  the  same  shutter  shaft.  Both  shutters  are  housed 
in  suitable  protective  guards,  and  means  are  provided  for  the  easy  removal  of 
these  guards  when  necessary  for  cleaning  purposes.  Attached  to  the  rear  shutter, 
between  the  source  of  illumination  and  the  aperture  plate,  are  specially  designed 
vanes,  the  function  of  which  is  to  create  a  partial  vacuum  in  and  around  the 
aperture  plate  and  remove  therefrom  at  high  velocity  the  heat  created  by  the 
illuminated  spot  and  keep  the  entire  rear  of  the  projector  cool  enough  to  touch 
with  the  fingers  even  when  using  high-intensity  arcs. 

To  set  the  shutters  in  exact  synchronism  and  in  proper  relation  to  the 
intermittent  movement,  a  shutter-setting  device  is  supplied  with  each  pair  of 
mechanisms.  With  this  device  the  shutters  may  be  set  in  an  extremely  simple 


Mar.,  1939] 


NEW  MOTION  PICTURE  APPARATUS 


327 


manner  and  with  a  greater  degree  of  accuracy  than  was  possible  with  earlier 
equipment. 

A  steadier  picture  is  obtained  both  vertically  and  laterally,  first,  through  new 
developments  in  intermittent  movement  design,  which  allow  far  greater  accuracy 
in  manufacture,  plus  a  hardened  and  ground  intermittent  sprocket  on  which  the 
64  radii  of  the  teeth  are  ground  to  extreme  accuracy;  and,  second,  by  the  addi- 
tion of  edge-guiding  in  the  film-trap,  which  maintains  the  film  in  a  constant  lateral 
position  and  does  not  allow  the  film  to  weave  slightly  as  in  former  designs. 

To  eliminate  the  possibility  of  improper  or  inadequate  lubrication,  the  equip- 
ment is  provided  with  the  Bijur  one-shot  oiling  system  (Fig.  3).  This  system  has 
proved  its  merit  in  the  finest  types  of  accounting  machines,  sewing  machines, 
electric  lamp  machinery,  etc.;  it  is  incorporated  in  the  best  American  trucks  and 
ambulances,  and  in  the  highest  grade  foreign  automobiles. 


FIG.  2.     Synchronized  front  and  rear  shutters. 


As  applied  to  the  new  projector,  this  system  consists  of  a  piston  pump  which 
delivers  at  each  impulse  a  metered  quantity  of  filtered  oil  to  a  distribution  system 
where,  by  means  of  meter  units,  this  measured  quantity  is  proportioned  to  each 
bearing  in  predetermined  quantities.  Check- valves  in  the  meter  units  prevent 
draining  the  oil  lines  between  "shots."  The  system  is  provided  with  a  double  set 
of  dense  felt  filters,  the  one  in  the  pump  being  the  denser  and  serving  to  filter  the 
reservoir  oil;  the  other  set  of  filters  is  in  the  individual  meter  units  and  is  for 
protection  of  the  units  against  chips  and  dirt  before  and  during  assembly.  This 
combination  of  filtering  assures  a  clean  supply  of  oil  at  every  bearing.  The 
Bijur  Company  advise  that  they  have  as  yet  found  no  indication  of  a  limit  to  the 
life  of  the  filters  when  used  with  a  straight  mineral  oil.  The  pump  develops  a 
pressure  of  35  pounds  per  square  inch. 

The  small  tubes  leading  from  the  meter  units  to  the  bearings  are  so  propor- 
tioned, as  to  ratio  of  wall  thickness  to  bore,  that  they  will  not  collapse  even  when 
bent  double  in  a  vise,  so  there  is  no  way  in  which  they  could  be  stopped  up  with- 
out a  type  of  handling  that  would  also  damage  the  mechanism  as  a  whole.  With 


328 


NEW  MOTION  PICTURE  APPARATUS        [J.  S.  M.  P.  E 


>ȣ 


.J3    f> 

'"S  5 


Mar.,  1939]  NEW  MOTION  PICTURE  APPARATUS  329 

a  lubricating  system  such  as  this  the  only  place  in  which  it  is  necessary  to  see  the 
oil  is  in  the  pump  unit,  and  a  sight-glass  is  provided  for  this  purpose. 

Such  bearings  as  are  not  reached  by  this  lubricating  system  are  the  ball  bearings 
that  carry  the  shutter-shaft  assembly  and  the  lower  bearing  of  the  oblique  shaft. 
These  are  of  the  sealed,  grease-packed,  dustproof  type,  and  do  not  need  attention 
for  a  number  of  years.  Oil-can  lubrication  is  required  only  to  fill  the  oil  reservoir 
of  the  Bijur  system  and  the  intermittent  movement,  which  will  be  discussed  later. 

Considering  fire  protection  of  paramount  importance,  the  equipment  is  sup- 
plied with  a  positive-acting  fire-shutter  between  the  rear  shutter  and  the  film 
aperture  (Fig.  4),  which  is  operated  through  a  centrifugally  operated  disk  mounted 
on  the  revolving  shutter- shaft.  When  the  projector  is  at  rest  this  disk  lies  at  an 


FIG.  5.     Automatic  fire-shutter  safety  trip. 


angle  with  relation  to  the  shutter-shaft  and  when  the  projector  is  in  operation, 
centrifugal  motion  straightens  up,  so  to  speak,  and  through  a  unique  linkage 
device  raises  the  fire-shutter  out  of  the  beam  of  light. 

It  is  well  known  that  on  a  properly  designed  projector  mechanism,  and  one  that 
is  kept  in  good  repair,  it  is  not  possible  under  normal  circumstances  to  burn  more 
than  one  frame  of  film  at  the  aperture  plate  should  a  splice  part  at  the  intermittent 
sprocket,  and  this,  incidentally,  is  the  only  time  it  is  possible  during  normal  op- 
eration of  a  projector  for  any  kind  of  fire  to  occur.  However,  to  eliminate  even 
this  cause  of  fire,  the  apparatus  is  provided  with  an  automatic  fire-shutter  safety 
trip  (Fig.  5),  which  operates  in  connection  with  the  automatic  fire-shutter  and 
instantaneously  drops  the  latter  should  a  splice  part  at  the  intermittent  sprocket. 
The  unit  is  operated  by  the  slightest  increase  in  size  of  the  upper  loop,  and  the 


330 


NEW  MOTION  PICTURE  APPARATUS        [J.  S.  M.  P.  E. 


action  is  instantaneous,  the  fire-shutter  dropping  before  the  film  has  a  chance  to 
ignite.  This  means  increased  protection  to  the  projectionist  and  theater  owner. 
The  device  is  simply  attached  and  may  readily  be  removed  and  replaced. 

A  newly  designed  film-gate  has  been  provided,  of  much  heavier  construction 
than  any  of  its  predecessors  and  readily  removable  by  simply  removing  two  thumb- 
nuts  and  sliding  it  from  its  two  supporting  studs  (Fig.  6).  This  gate  is  now  formed 
from  one  heavy  steel  stamping  which  supports  along  its  entire  length  pressure 
pads  of  new  design,  the  tension  on  all  of  which,  through  self-equalizing  cone 


II 


%i f 

FIG.  6.     Film  gate:    (left)  rear  view;    (right)  front  view. 

springs,  is  readily  adjusted  by  the  projectionist  while  the  projector  is  in  operation. 
The  base  of  the  gate  supports  the  intermittent  sprocket  tension  shoe  which  also 
is  provided  with  an  adjustable  tension  unit.  With  this  type  of  gate,  and  due  to 
the  proper  placement  of  tension  shoes,  it  has  been  found  possible  to  lessen  the 
tension  considerably  and  at  the  same  time  maintain  a  much  steadier  picture  than 
heretofore. 

The  gate  is  mounted  on  an  opening  and  closing  unit  of  entirely  new  design  and 
is  locked  solidly  in  both  its  open  and  closed  position.  Provision  is  made  for 
removing  entirely  any  lateral  displacement  of  the  gate  due  to  wear,  and  this 
combination  definitely  prevents  any  "jiggling"  of  the  gate  such  as  was  some- 
times evident  in  earlier  models.  The  entire  gate  is  provided  with  a  very  simple 


Mar.,  1939]  NEW  MOTION  PICTURE  APPARATUS 


331 


332  NEW  MOTION  PICTURE  APPARATUS        [J.  S.  M.  p.  E. 

lateral  adjustment  by  means  of  which  it  may  be  correctly  aligned  with  the  runners 
on  the  film-trap. 

The  film-trap  is  of  completely  new  and  original  design  and  may  now  be  removed 
entirely  for  cleaning  purposes  as  may  the  gate  (Fig.  7).  This  is  accomplished 
simply  by  the  removal  of  two  screws,  whereupon  the  film-trap  may  be  slid  off 
toward  the  projectionist.  The  film-trap  is  provided  with  the  conventional 
lateral  guide-rollers,  but,  in  addition  thereto,  edge-guiding  channels  have  been 
added  which,  in  cooperation  with  the  lateral  guide-rollers,  maintain  the  edge  of 
the  film  steadily  against  the  edge  of  the  guide,  thus  eliminating  all  sidesway  during 
film  travel.  The  film-handling  parts  of  both  the  gate  and  film-trap  are  readily 
removable  for  replacement  when  wear  becomes  apparent,  and  it  is  no  longer 
necessary  to  tear  down  the  entire  projector  when  such  is  the  case. 


FIG.  8.     Framing  lamp  and  spot  sight  box. 


As  in  the  case  of  previous  film-trap  and  gate  design  in  Simplex  projectors,  the 
emulsion  side  is  always  toward  the  film-trap  runners;  thus,  variations  in  film 
thickness,  as  between  standard  production  film  and  newsreel  stock,  will  not  affect 
the  definition  of  the  projected  picture;  thus,  it  is  not  necessary  to  re-focus  the 
lens  to  compensate  for  the  difference  in  the  position  of  the  emulsion  as  is  the  case 
with  the  old  Powers  projector  mechanism  and  other  equipment  of  similar  design. 

An  ingenious  framing  lamp  and  spot  sight  box  assembly  is  mounted  between 
the  rear  shutter  and  the  film-trap  (Fig.  8).  A  small  incandescent  lamp  of  the 
bayonet  type  is  pivotally  mounted  in  this  assembly  and  operable  only  when  the 
fire-shutter  is  raised  manually  during  the  process  of  threading  the  film  into  the 
projector;  thus,  a  strong,  direct  beam  of  light  is  projected  through  the  aperture 
plate  by  means  of  which  the  projectionist  may  accurately  select  the  frame  of  film 
to  be  placed  over  the  latter.  The  lamp  is  lighted  through  a  small  mercoid  switch 
and  extinguished  upon  release  of  the  fire-shutter  lever  which  automatically  takes 
the  framing  lamp  assembly  out  of  the  path  of  the  projection  lamp  beam  and  ex- 
tinguishes it  at  the  same  time. 


Mar.,  1939  J 


NEW  MOTION  PICTURE  APPARATUS 


333 


The  spot  sight  box  in  which  this  assembly  is  mounted  is  provided  with  a  number 
of  air-cooled  fins  for  rapid  dissipation  of  heat  from  the  projection  lamp-beam.  A 
highly  polished  nickel-plated  copper  baffle,  together  with  two  additional  copper 
baffles,  form  part  of  the  heat-reflecting  unit  in  this  assembly,  and  this  helps 
further  to  cool  the  rear  of  the  mechanism.  The  entire  assembly  is  readily  remov- 
able for  cleaning,  and  engages  the  electrical  circuit  by  means  of  pin  plugs  when  in 
its  proper  operating  position. 

The  intermittent  movement  in  this  equipment  is  of  the  conventional  double- 
bearing  type,  similar  to  that  provided  in  earlier  Simplex  mechanisms  (Fig.  9). 
It  has  been  greatly  improved,  however,  as  to  oil  sealing,  and  is  provided  with 


FIG.  9.     Cut-away  views  of  intermittent  movement. 

means  whereby  it  is  impossible  to  insert  a  greater  quantity  of  oil  than  is  required 
for  proper  operation.  This  eliminates  the  possibility  of  the  projectionist's  over- 
filling the  oil-well  and  thus  allowing  the  lubricant  to  spill  over;  and  at  the  same 
time  makes  it  possible  at  all  times  for  him  to  see  through  the  sight  glasses  in  the 
movement,  that  the  oil  level  is  correctly  maintained. 

The  entire  movement  is  of  considerably  heavier  construction  throughout,  and 
its  design,  as  before  stated,  allows  for  far  more  accurate  construction  than  was 
possible  in  any  previous  movement.  In  addition  to  being  leakproof  it  is  practi- 
cally bindproof,  since  properly  designed  spiral  grooves  in  connection  with  oil 
channels  are  provided  in  the  revolving  shafts  and  bearings  which  carry  oil  to  all 
parts  needing  constant  lubrication.  In  addition  to  this,  perfect  lubrication  be- 
tween the  cam  ring  and  the  star  radius  is  assured,  since  oil  is  forced  through  two 
channels  in  the  cam  forming  a  cushion  of  oil  between  the  star  radius  and  the  cam 


334 


NEW  MOTION  PICTURE  APPARATUS        [J.  S.  M.  P.  E. 


ring  when  they  lock  together  while  the  picture  is  being  projected.  This  also 
cushions  the  blow  between  the  two  units  and  makes  for  a  quiet-running  unit. 
The  movement  may  be  lubricated  from  either  the  non-operating  or  operating  side. 

One  of  the  important  features  in  connection  with  this  intermittent  movement 
is  the  fact  that  it  may  be  readily  removed  for  cleaning,  or  parts  replaced  with- 
out disturbing  any  of  the  major  parts  of  the  projector  mechanism  or  sound- 
reproducing  unit — and  this  from  the  operating  side  of  the  mechanism. 

Positive  synchronism,  without  backlash  or  lost  motion,  is  assured  between  the 
intermittent  movement  and  the  revolving  shutters  when  framing  by  the  uniquely 
designed  assembly  now  performing  this  function.  The  shutter-shaft  passes 
through  an  assembly  similar  to  a  cylinder  and  piston  in  automobile  design  (Fig. 
10),  and  fastened  to  the  piston  through  a  ball-race  is  the  shutter  driving  gear  by 
means  of  which  the  shutter-shaft  is  driven  through  a  woodruff  key.  Attached 


? 

i 
m  J 


FIG.  10.     New  ring  type  governor. 

also  to  the  piston  is  a  pivoted  arm,  the  lower  end  of  which  fetches  up  solidly 
against  a  plunger  pin  operated  by  the  framing  cam  of  the  intermittent  movement 
assembly. 

The  entire  assembly  is  held  under  substantial  tension  by  means  of  a  heavy 
coiled  spring,  one  end  of  which  is  held  under  tension  by  a  collar  on  the  shutter- 
shaft  and  revolving  with  it,  and  the  other  end  of  which  fetches  up  solidly  against 
the  ball-race  attached  to  the  gear.  This  spring  performs  two  functions :  it  forces 
the  piston  rearward  at  all  times,  and  at  the  same  time  removes  any  slight  end- 
play  in  the  shutter  shaft  and  framing  device  synchronizing  assembly. 

In  operation,  when  the  framing  handle  is  turned  in  one  direction,  the  intermit- 
tent movement,  complete  with  its  framing  cam,  revolves  in  its  housing,  forces  the 
plunger  against  the  pivoted  arm,  which  in  turn  moves  the  piston  and  gear  as- 
sembly forward  against  the  spring  compression,  thus  revolving  the  spiral  gear  and 
shutter-shaft  the  exact  amount  necessary  to  maintain  synchronism  between  the 
intermittent  movement  and  the  revolving  shutters.  When  the  framing  handle  is 
turned  in  the  opposite  direction,  the  entire  assembly  performs  exactly  the  same 
function,  except  that  the  compression  spring  forces  the  piston  and  gear  assembly 


Mar.,  1939]  NEW  MOTION  PICTURE  APPARATUS  335 

rearward,  and  thus  the  same  synchronism  is  obtained  with  any  position  of  the 
framing  handle  and  intermittent  sprocket. 

All  bevel  gears  have  been  eliminated  and,  as  a  matter  of  fact,  the  number  of 
gears  has  been  greatly  reduced.  Spiral  gears  alone  now  form  the  driving  equip- 
ment, and  thus  the  noise-level  during  operation  has  been  tremendously  reduced. 
The  face  of  the  main  drive  gears  has  been  increased  in  cross-section  as  a  protection 
against  the  high  starting  torque  of  the  modern  sound-head,  and  these  gears  now 
operate  on  hardened  and  ground  studs  rigidly  attached  to  the  center  frame.  The 
area  of  the  bearings  of  the  gears  that  revolve  upon  these  studs  has  also  been 
greatly  increased,  and  lubrication  is  provided  through  the  Bijur  one-shot  oiling 
system  to  the  center  of  the  studs,  forcing  in  clean  oil  all  the  time  and  washing  any 
dirty  lubricant  out  on  the  non-operating  side  of  the  projector  only;  thus  re- 
bushing  of  the  main  frame  in  this  connection  has  been  eliminated,  and  much 
longer  life  is  assured  and  cost  of  maintenance  reduced. 

Wherever  the  finest  accuracy  is  not  required  non-scoring  bearing  material  is 
used ;  but  where  extreme  accuracy  is  required,  and  this  material  can  not  be  relied 
upon  satisfactorily,  the  type  of  bearings  best  suited  for  their  proper  function,  such 
as  ball  bearings  on  the  shutter  shaft,  and  burnished  cast-iron  bearings  for  hard- 
ened and  ground  shafts,  are  used. 

Lens  focusing  is  accomplished  from  either  inside  or  outside  the  mechanism, 
making  it  possible  at  all  times  to  control  readily  the  definition  of  the  projected 
picture. 

The  mechanism  is  designed  to  fit  any  existing  standard  sound-reproducing 
apparatus.  The  interior  is  finished  in  pearl-gray  enamel,  to  facilitate  ob- 
servation of  the  film  travel.  This  light  interior  also  lends  itself  to  cleanliness,  and 
is  brightly  illuminated  by  the  additional  threading  lamp  provided  in  the  upper 
right-hand  corner  of  the  operating  side  of  the  mechanism.  This  latter  lamp 
eliminates  the  need  for  the  old-fashioned  trouble-lamp  heretofore  necessary  for 
checking  during  threading  operations. 


1939  SPRING  CONVENTION 

SOCIETY  OF  MOTION  PICTURE  ENGINEERS 


HOLLYWOOD  ROOSEVELT  HOTEL 
HOLLYWOOD,   CALIFORNIA 
APRIL  17th-21st,  INCLUSIVE 

Officers  and  Committees  in  Charge 

E.  A.  WILLIFORD,  President 

N.  LEVINSON,  Executive  Vice-P resident 

W.  C.  KUNZMANN.  Convention  Vice-President 

J.  I.  CRABTREE,  Editorial  Vice-President 

L.  L.  RYDER,  Chairman,  Pacific  Coast  Section 

H.  G.  TASKER,  Chairman,  Local  Arrangements  Committee 

J.  HABER,  Chairman,  Publicity  Committee 

Pacific  Coast  Papers  Committee 

L.  A.  AICHOLTZ,  Chairman 

O.  O.  CECCARINI  C.  N.  BATSEL  W.  H.  ROBINSON,  JR 

G.  CHAMBERS  W.  A.  MUELLER  C.  R.  SAWYER 

L.  D.  GRIGNON  H.  G.  TASKER  R.  TOWNSEND 

Reception  and  Local  Arrangements 

H.  G.  TASKER,  Chairman 

N.  LEVINSON  G.  F.  RACKETT  E.  HUSE 

K.  F.  MORGAN  H.  W.  MOYSE  L.  L.  RYDER 

P.  MOLE  W.  MILLER  J.  O.  AALBERG 

A.  M.  GUNDELFINGER       J.  A.  BALL  R.  H.  McCULLOUGH 

H.  W.  REMERSCHIED       W.  A.  MUELLER  J.  M.  NICKOLAUS 

G.  S.  MITCHELL  C.  L.  LOOTENS  E.  H.  HANSEN 

Registration  and  Information 

W.  C.  KUNZMANN,  Chairman 

S.  HARRIS  C.  W.  HANDLEY 

E.  R.  GEIB  W.  R.  GREENE 

Hotel  and  Transportation 

G.  A.  CHAMBERS,  Chairman 

W.  C.  HARCUS  C.  DUNNING  W.  E.  THEISEN 

G.  HOUGH  E.  C.  RICHARDSON  D.  P.  LOYE 

B.  KREUZER  K.  STRUSS  O.  O.  CECCARINI 
H.  W.  REMERSCHIED       J.  C.  BROWN                          C.  J.  SPAIN 

336 


1939  SPRING  CONVENTION  337 

Convention  Projection 

H.  GRIFFIN,  Chairman 

J.  O.  AALBERG  H.  A.  STARKE  J.  DURST 

C.  W.  HANDLEY  L.  E.  CLARK  R.  H.  McCuLLOucn 

W.  F.  RUDOLPH  M.  S.  LESHING  H.  C.  SILENT 

J.  M.  NICKOLAUS  A.  F.  EDOUART  H.  I.  REISKIND 

J.  K.  HILLIARD  I.  SERRURIER  W.  W.  LINDSAY,  JR. 

Officers  and  Members  of  Los  Angeles  Projectionists  Local  No.  150 

Banquet  and  Dance 

N.  LEVINSON,  Chairman 

H.  T.  KALMUS  G.  S.  MITCHELL  G.  F.  RACKETT 

E.  HUSE  P.  MOLE  J.  O.  AALBERG 

L.  L.  RYDER  H.  G.  TASKER  K.  F.  MORGAN 

C.  DUNNING  W.  MILLER  H.  W.  MOYSE 

G.  A.  MITCHELL  R.  H.  McCuLLOUGH  J.  L.  COURCIER 

Ladies'  Reception  Committee 

MRS.  N.  LEVINSON,  Hostess 

assisted  by 

MRS.  E.  C.  RICHARDSON  MRS.  P.  MOLE  MRS.  E.  HUSE 

MRS.  G.  F.  RACKETT       MRS.  C.  W.  HANDLEY  MRS.  L.  L.  RYDER 

MRS.  H.  W.  MOYSE         MRS.  K.  F.  MORGAN  MRS.  J.  O.  AALBERG 

MRS.  H.  G.  TASKER        MRS.  W.  MILLER  MRS.  R.  H.  MCCULLOUGH 

MRS.  L.  E.  CLARK  MRS.  A.  M.  GUNDELFINGER  MRS.  C.  DUNNING 

Publicity 

J.  HABER,  Chairman 

L.  A.  AICHOLTZ  W.  R.  GREENE  S.  HARRIS 

W.  A.  MUELLER  G.  CHAMBERS  A.  M.  GUNDELFINGER 

New  Equipment  Exhibit 
J.  G.  FRAYNE,  Chairman 

P.  MOLE  C.  R.  DAILY  H.  W.  REMERSCHIED 

J.  DURST  S.  HARRIS  C.  N.  BATSEL 

O.  F.  NEU 

Headquarters 

Headquarters  of  the  Convention  will  be  the  Hollywood  Roosevelt  Hotel,  where 
excellent  accommodations  are  assured.  A  reception  suite  will  be  provided  for  the 
Ladies'  Committee,  and  an  excellent  program  of  entertainment  will  be  arranged 
for  the  ladies  who  attend  the  Convention. 

Special  hotel  rates,  guaranteed  to  SMPE  delegates,  European  plan,  will  be  as 
follows : 

One  person,  room  and  bath  $  3 . 50 

Two  persons,  double  bed  and  bath  5 . 00 

Two  persons,  twin  beds  and  bath  6 . 00 

Parlor  suite  and  bath,  1  person  8 . 00 

Parlor  suite  and  bath,  2  persons  12.00 


338  1939  SPRING  CONVENTION  [j.  s.  M.  p.  E. 

Indoor  and  outdoor  garage  facilities  adjacent  to  the  Hotel  will  be  available 
to  those  who  motor  to  the  Convention. 

Members  and  guests  of  the  Society  will  be  expected  to  register  immediately 
upon  arriving  at  the  Hotel.  Convention  badges  and  identification  cards  will 
be  supplied  which  will  be  required  for  admittance  to  the  various  sessions,  the 
studios,  and  several  Hollywood  motion  picture  theaters. 

Railroad  Fares 

The  following  table  lists  the  railroad  fares  and  Pullman  charges : 

Railroad 

Fare  Pullman 

City  (round  trip)  (one  way) 

Washington  $132.20  $22.35 

Chicago  90.30  16.55 

Boston  147.50  23.65 

Detroit  106.75  19.20 

New  York  139.75  22.85 

Rochester  124.05  20.50 

Cleveland  110.00  19.20 

Philadelphia  135.50  22.35 

Pittsburgh  117.40  19.70 

The  railroad  fares  given  above  are  for  round  trips,  sixty-day  limits.  Arrange- 
ments may  be  made  with  the  railroads  to  take  different  routes  going  and  coming, 
if  so  desired,  but  once  the  choice  is  made  it  must  be  adhered  to,  as  changes  in  the 
itinerary  may  be  effected  only  with  considerable  difficulty  and  formality.  Dele- 
gates should  consult  their  local  passenger  agents  as  to  schedules,  rates,  and  stop- 
over privileges. 

San  Francisco  Fair 

On  February  18,  1939,  the  Golden  Gate  Exposition  opened  at  San  Francisco, 
an  overnight  trip  from  Hollywood.  The  exposition  will  last  throughout  the  sum- 
mer so  that  opportunity  will  be  afforded  the  eastern  members  of  the  Society  to 
take  in  this  attraction  on  their  Convention  trip.  Special  arrangements  have  been 
made  with  the  Hotel  Empire  at  San  Francisco  for  Convention  delegates  visiting 
the  Fair,  at  the  following  daily  rates : 

One  person,  room  and  bath  $3 . 50 

Two  persons,  double  bed  and  bath  5 . 00 

Two  persons,  twin  beds  and  bath  6 . 00 

Suites: 

Parlor  and  bedroom  for  two  persons  8 . 00  and  up 

Two  large  bedrooms,  each  with  private  bath  and  a  living 

room;  for  four  persons  16.00 

Reservations  can  be  made  either  by  writing  directly  to  the  Hotel  Empire  or  by 
addressing  Mr.  W.  C.  Kunzmann,  Convention  Vice-President,  Box  6087,  Cleve- 
land, Ohio. 


Mar.,  1939]  1939  SPRING  CONVENTION  339 

Technical  Sessions 

The  Hollywood  meeting  always  offers  our  membership  an  opportunity  to  be- 
come better  acquainted  with  the  studio  technicians  and  production  problems,  and 
arrangements  will  be  made  to  visit  several  of  the  studios.  The  Local  Papers 
Committee  under  the  chairmanship  of  Mr.  L.  A.  Aicholtz  is  collaborating  closely 
with  the  General  Papers  Committee  in  arranging  the  details  of  the  program. 
Complete  details  of  the  program  will  be  published  in  a  later  issue  of  the  JOURNAL. 

Studio  Visits 

On  the  afternoon  of  Tuesday,  April  18th,  Paramount  Pictures,  Inc.,  will  act  as 
hosts  of  the  Convention  at  their  Hollywood  Studio.  The  program  will  be  in 
charge  of  Messrs.  L.  L.  Ryder  and  H.  G.  Tasker.  On  the  afternoon  of  Thursday, 
April  20th,  the  delegates  of  the  Convention  will  be  entertained  at  the  studio  of 
Warner  Brothers  First  National,  Inc.,  at  Burbank.  The  program  of  the  afternoon 
will  be  under  the  supervision  of  Mr.  N.  Levinson. 

Semi- Annual  Banquet  and  Dance 

The  Semi- Annual  Banquet  of  the  Society  will  be  held  at  the  Hotel  on  Thursday, 
April  20th.  Addresses  will  be  delivered  by  prominent  members  of  the  industry, 
followed  by  dancing  and  entertainment.  Tables  reserved  for  8,  10,  or  12  persons; 
tickets  obtainable  at  the  registration  desk. 

New  Equipment  Exhibit 

An  exhibit  of  newly  developed  motion  picture  equipment  will  be  held  in  the 
Bombay  and  Singapore  Rooms  of  the  Hotel,  on  the  mezzanine.  Those  who  wish 
to  enter  their  equipment  in  this  exhibit  should  communicate  as  early  as  possible 
with  the  general  office  of  the  Society  at  the  Hotel  Pennsylvania,  New  York,  N.  Y. 

Motion  Pictures 

At  the  time  of  registering,  passes  will  be  issued  to  the  delegates  to  the  Conven- 
tion, admitting  them  to  the  following  motion  picture  theaters  in  Hollywood,  by 
courtesy  of  the  companies  named:  Grauman's  Chinese  and  Egyptian  Theaters 
(Fox  West  Coast  Theaters  Corp.),  Warner's  Hollywood  Theater  (Warner  Brothers 
Theaters,  Inc.),  Pantages  Hollywood  Theater  (Rodney  Pantages,  Inc.).  These 
passes  will  be  valid  for  the  duration  of  the  Convention. 

Ladies'  Program 

An  especially  attractive  program  for  the  ladies  attending  the  Convention  is 
being  arranged  by  Mrs.  N.  Levinson,  hostess,  and  the  Ladies'  Committee.  A 
suite  will  be  provided  in  the  Hotel,  where  the  ladies  will  register  and  meet  for 
the  various  events  upon  their  program.  Further  details  will  be  published  in  a 
succeeding  issue  of  the  JOURNAL. 

Points  of  Interest 

En  route:  Boulder  Dam,  Las  Vegas,  Nevada;   and  the  various  National  Parks. 
Hollywood  and  vicinity:    Beautiful  Catalina  Island;    Zeiss  Planetarium;    Mt. 


340  1939  SPRING  CONVENTION 

Wilson  Observatory;  Lookout  Point,  on  Lookout  Mountain;  Huntington  Li- 
brary and  Art  Gallery  (by  appointment  only) ;  Palm  Springs,  Calif. ;  Beaches  at 
Ocean  Park  and  Venice,  Calif.;  famous  old  Spanish  missions;  Los  Angeles  Mu- 
seum (housing  the  SMPE  motion  picture  exhibit);  Mexican  village  and  street, 
Los  Angeles. 

In  addition,  numerous  interesting  side  trips  may  be  made  to  various  points 
throughout  the  West,  both  by  railroad  and  bus.  Among  the  bus  trips  available 
are  those  to  Santa  Barbara,  Death  Valley,  Agua  Caliente,  Laguna,  Pasadena, 
and  Palm  Springs,  and  special  tours  may  be  made  throughout  the  Hollywood 
area,  visiting  the  motion  picture  and  radio  studios. 


SOCIETY  SUPPLIES 

The  following  are  available  from  the  General  Office  of  the  Society,  at  the  prices 
noted.  Orders  should  be  accompanied  by  remittances. 

Aims  and  Accomplishments. — An  index  of  the  Transactions  from  October, 
1916,  to  December,  1929,  containing  summaries  of  all  articles,  and  author  and 
classified  indexes.  One  dollar  each. 

Journal  Index. — An  index  of  the  JOURNAL  from  January,  1930,  to  December, 
1935,  containing  author  and  classified  indexes.  One  dollar  each. 

SMPE  Standards.— The  revised  edition  of  the  SMPE  Standards  and  Recom- 
mended Practice  was  published  in  the  March,  1938,  issue  of  the  JOURNAL,  copies 
of  which  may  be  obtained  for  one  dollar  each. 

Membership  Certificates. — Engrossed,  for  framing,  containing  member's  name, 
grade  of  membership,  and  date  of  admission.  One  dollar  each. 

Lapel  Buttons. — The  insignia  of  the  Society,  gold  filled,  with  safety  screw  back. 
One  dollar  each. 

Journal  Binders. — Black  fabrikoid  binders,  lettered  in  gold,  holding  a  year's 
issue  of  the  JOURNAL.  Two  dollars  each.  Member's  name  and  the  volume 
number  lettered  in  gold  upon  the  backbone  at  an  additional  charge  of  fifty  cents 
each. 

Test- Films. — See  advertisement  in  this  issue  of  the  JOURNAL. 


SOCIETY  ANNOUNCEMENTS 


HOLLYWOOD  CONVENTION 

As  outlined  in  the  preceding  section  of  this  issue,  and  also  as  announced  on  the 
inside  front  cover,  the  next  Convention  of  the  Society  will  be  held  on  April  17th- 
21st,  inclusive,  at  Hollywood,  Calif.,  with  headquarters  at  the  Hollywood 
Roosevelt  Hotel.  Full  details  concerning  the  program  will  be  published  in  the 
next  issue  of  the  JOURNAL. 

ATLANTIC  COAST  SECTION 

At  a  meeting  held  at  the  Hotel  McAlpin,  New  York,  N.  Y.,  on  Wednesday, 
February  15th,  Mr.  Clinton  P.  Veber,  Research  Associate  of  the  Department  of 
Biophotography  at  Rutgers  University,  presented  a  paper  describing  "The 
Time  Telescope." 

A  demonstration  of  the  instrument  accompanied  the  paper,  which  discussed 
the  use  and  history  of  time-lapse  photography,  and  also  the  control,  design,  and 
operation  of  the  new  equipment  used  at  Rutgers  University.  Films  were  shown 
illustrating  the  use  of  machines  in  producing  spectacular  pictures  showing  the 
life  histories  of  plants  and  other  botanical  subjects. 

MID-WEST  SECTION 

The  first  meeting  of  the  year  was  held  on  January  24th  in  the  meeting  rooms 
of  the  Western  Society  of  Engineers  at  Chicago.  Mr.  E.  F.  Lowry,  Research 
Director  of  the  Continental  Electric  Company,  Chicago,  gave  an  interesting 
talk  on  "The  Theory  and  Operation  of  Rectifier  Tubes  and  Cathodes."  Briefly 
tracing  the  history  of  vacuum-tubes  from  the  beginning  to  the  present  day,  the 
paper  turned  to  the  subject  of  rectifier  tubes,  in  particular,  those  of  the  mercury- 
vapor  type. 

Following  the  talk,  Mr.  J.  G.  Black  gave  a  demonstration  of  a  6-phase  mercury- 
vapor  rectifier. 

AMENDMENTS 

At  the  Detroit  Convention  of  the  Society,  on  October  31,  1938,  the  following 
amendment  of  the  Constitution  was  proposed: 

Article  IV,  Officers 

It  is  proposed  that  the  term  of  office  of  the  Executive  Vice-P resident  be  extended  to 
two  years,  in  view  of  the  fact  that  the  terms  of  all  the  other  vice-presidents  are  two  years. 
Original  wording: 

The  officers  of  the  Society  shall  be  a  President,  a  Past-President,  an  Executive 
Vice-President,  an  Engineering  Vice-President,  an  Editorial  Vice- President,  a 
Financial  Vice-President,  a  Convention  Vice-President,  a  Secretary,  and  a  Trea- 
surer. 

341 


342  SOCIETY  ANNOUNCEMENTS  [J.  S.  M.  P.  E. 

The  term  of  office  of  the  President  and  Past- President  shall  be  two  years;  of 
the  Engineering,  Editorial,  Financial,  and  Convention  Vice-Presidents,  two  years; 
and  of  the  Executive  Vice-President,  Secretary,  and  Treasurer,  one  year.  Of  the 
Engineering,  Editorial,  Financial,  and  Convention  Vice-Presidents,  two  shall  be 
elected  alternately  each  year  or  until  their  successors  are  chosen.  The  Presi- 
dent shall  not  be  immediately  eligible  to  succeed  himself  in  office. 
Proposed  wording: 

The  officers  of  the  Society  shall  be  a  President,  a  Past-President,  an  Executive 
Vice-President,  an  Engineering  Vice-President,  an  Editorial  Vice-President,  a 
Financial  Vice-President,  a  Convention  Vice-President,  a  Secretary,  and  a 
Treasurer. 

The  term  of  office  of  the  President,  the  Past-President,  the  Executive  Vice- 
President,  the  Engineering  Vice-President,  the  Editorial  Vice-President,  the 
Financial  Vice-President,  and  the  Convention  Vice-President  shall  be  two  years, 
and  the  Secretary  and  the  Treasurer  one  year.  Of  the  Engineering,  Editorial, 
Financial,  and  Convention  Vice-Presidents,  two  shall  be  elected  alternately  each 
year,  or  until  their  successors  are  chosen.  The  President  shall  not  be  imme- 
diately eligible  to  succeed  himself  in  office. 

Adhering  to  the  procedure  for  voting  upon  amendments  of  the  Constitution, 
voting  ballots  were  mailed  to  the  voting  membership  shortly  after  the  Conven- 
tion, and  a  recent  count  of  the  ballots  by  the  Secretary  indicated  practically 
unanimous  approval  of  this  amendment. 

ADMISSIONS  COMMITTEE 

At  a  recent  meeting  of  the  Admissions  Committee  at  the  General  Office  of  the 
Society,  the  following  applicants  for  membership  were  admitted  to  the  Associate 
grade: 

ACKLEY,  A.  COHEN,  C. 

270  North  Michigan,  1821  Roselyn  St., 

Chicago,  111.  Philadelphia,  Penna. 

BAILWARD,  P.  COLEMAN,  J.  A. 

6  Pall  Mall,  13553  Artesian  Ave., 

London,  England.  Detroit,  Mich. 

BARUA,  P.  C.  COLTON,  H.  C. 

14  Ballygunge  Circular  Rd.,  119-40  Union  Turnpike, 

Calcutta,  India.  Kew  Gardens,  N.  Y. 

BERGSTEDT,  F.  H.  COOPER,  H.  G. 

2780  Dewey  Ave.,  1015  N.  Edinburgh, 

Rochester,  N.  Y.  Los  Angeles,  Calif. 

BIRCH,  H.  DANUFF,  I.  R. 

609  Stratford  PL,  1050  Anderson  Ave., 

Chicago,  111.  Bronx,  N.  Y. 

BRADSHAW,  C.  H.  FEDERICI,  M. 

12708  Littlefield  St.,  Corti  12, 

Detroit,  Mich.  Milan,  Italy. 

BRANDT,  J.  S.  GRANT,  S. 

448  Lincoln  Ave.,  35  E.  Wacker  Dr., 

Orange,  N.  J.  Chicago,  111. 


Mar.,  1939] 


SOCIETY  ANNOUNCEMENTS 


343 


HALL,  F. 

119  LeRoy  St., 

New  York,  N.  Y. 
HALL,  H.  W. 

Beeville,  Tex. 
JAPIKSE,  A.  B. 
22  Tulpweg, 

Wassenaar,  Holland. 
JENNINGS,  B.  D. 
245  W.  55th  St., 

New  York,  N.  Y. 
KENNEDY,  F.  M. 
231  S.  Witmer  St., 

Los  Angeles,  Calif. 
KENNEDY,  J.  D. 

14668  Abington  Rd., 
Detroit,  Mich. 

KNIFFEN,  L.  D. 

Kincardine,  Ontario,  Canada. 

KRAMER,  L. 
1513  Field  St., 
Detroit,  Mich. 

KUTTNAUER,  L.  V. 

1223  S.  Wabash  Ave., 

Chicago,  111. 
MENLEY,  F.  A. 
931  Ogden  Ave.,  S.  E., 
Grand  Rapids,  Mich. 
MORANZ,  J. 
Breslin  Bldg., 

Louisville,  Ky. 
MORELOCK,  O.  J. 

614  Frelinghuysen  Ave., 

Newark,  N.  J. 
MUDGE,  M.  L. 

P.  O.  Box  41,  Linwood  Station, 
Detroit,  Mich. 

NlLLESEN,  H.  A. 

N.  V.  Philips, 

Eindhoven,  Holland. 


OLIN,  N.  E. 

126  W.  73rd  St., 
New  York,  N.  Y. 

OWNBEY,  L.  C. 

255  Golden  Gate  Ave., 
San  Francisco,  Calif. 
PARK,  W.  C. 
278  N.  Fulton  Ave., 
Mt.  Vernon,  N.  Y. 
POLLOCK,  J.  R.,  JR. 
590  HAMILTON  St., 
Vancouver,  B.  C. 
POLLACK,  S.  M. 

Box  4—969  Hoe  Ave., 
New  York,  N.  Y. 

SCHOLES,  J.  J. 

100  Gibbs  St., 

Rochester,  N.  Y. 
SONKIN,  D. 

12  Dongan  PI., 

New  York,  N.  Y. 
THOMPSON,  F.  E. 

932  Collingwood  Ave., 

Detroit,  Mich. 
THOMPSON,  J. 

4552  Camellis  Ave., 

N.  Hollywood,  Calif. 
TOENNIES,  J.  F. 
404  E.  55th  St., 

New  York,  N.  Y. 
VANDERFORD,  H.  L. 
195  Broadway, 

New  York,  N.  Y. 
VOELKER,  C.  M. 
66  Sibley  St., 

Detroit,  Mich. 
WALL,  C.  R. 
39  Nassau  Ave., 

Malverne,  N.  Y. 
WOLFE,  H. 

2719  Hyperion  Ave., 
Los  Angeles,  Calif. 


In  addition  the  following  applicant  was  approved  by  the  Board  of  Governors 
for  transfer  from  the  Active  grade  to  the  grade  of  Fellow : 

LINDSAY,  W.  W.,  JR. 
6625  Romaine  St., 
Hollywood,  Calif. 


CONSTITUTION  AND  BY-LAWS 

OF  THE 
SOCIETY  OF  MOTION  PICTURE  ENGINEERS* 

CONSTITUTION 

Article  I 

Name 

The  name  of  this  association  shall  be  SOCIETY  OF  MOTION  PICTURE 
ENGINEERS. 

Article  II 

Object 

Its  objects  shall  be:  Advancement  in  the  theory  and  practice  of  motion  picture 
engineering  and  the  allied  arts  and  sciences,  the  standardizationof  the  equipment, 
mechanisms,  and  practices  employed  therein,  the  maintenance  of  a  high  profes- 
sional standing  among  its  members,  and  the  dissemination  of  scientific  knowledge 
by  publication. 

Article  III 
Eligibility 

Any  person  of  good  character  may  be  a  member  in  any  class  for  which  he  is 
eligible. 

Article  IV 

Officers 

The  officers  of  the  Society  shall  be  a  President,  a  Past-President,  an  Executive 
Vice-President,  an  Engineering  Vice- President,  an  Editorial  Vice-President,  a 
Financial  Vice-President,  a  Convention  Vice-President,  a  Secretary,  and  a  Trea- 
surer. 

The  term  of  office  of  the  President,  the  Past-President,  the  Executive  Vice- 
President,  the  Engineering  Vice-President,  the  Editorial  Vice-President,  the 
Financial  Vice-President,  and  the  Convention  Vice-President  shall  be  two  years, 
and  the  Secretary  and  the  Treasurer  one  year.  Of  the  Engineering,  Editorial, 
Financial,  and  Convention  Vice-Presidents,  two  shall  be  elected  alternately  each 
year,  or  until  their  successors  are  chosen.  The  President  shall  not  be  immedi- 
ately eligible  to  succeed  himself  in  office. 

Article  V 

Board  of  Governors 

The  Board  of  Governors  shall  consist  of  the  President,  the  Past-President,  the 
five  Vice-Presidents,  the  Secretary,  the  Treasurer,  the  Section  Chairmen,  and  five 
elected  Governors.  Two,  and  three,  of  the  Governors  shall  be  elected  alternately 
each  year  to  serve  for  two  years. 

*  Corrected  to  January  1,  1939. 
344 


CONSTITUTION  AND  BY-LAWS 


345 


Article  VI 

Meetings 

There  shall  be  an  annual  meeting,  and  such  other  meetings  as  stated  in  the  By- 
Laws. 

Article  VII 

Amendments 

This  Constitution  may  be  amended  as  follows :  Amendments  shall  be  approved 
by  the  Board  of  Governors,  and  shall  be  submitted  for  discussion  at  any  regular 
members'  meeting.  The  proposed  amendment  and  complete  discussion  then  shall 
be  submitted  to  the  entire  Active,  Fellow,  and  Honorary  membership,  together 
with  letter  ballot  as  soon  as  possible  after  the  meeting.  Two-thirds  of  the  vote 
cast  within  sixty  days  after  mailing  shall  be  required  to  carry  the  amendment. 

BY-LAWS 
By-Law  I 

Membership 

Sec.  1. — The  membership  of  the  Society  shall  consist  of  Honorary  members, 
Fellows,  Active  members,  Associate  members,  and  Sustaining  members. 

An  Honorary  member  is  one  who  has  performed  eminent  services  in  the  ad- 
vancement of  motion  picture  engineering  or  in  the  allied  arts.  An  Honorary 
member  shall  be  entitled  to  vote  and  to  hold  any  office  in  the  Society. 

A  Fellow  is  one  who  shall  not  be  less  than  thirty  years  of  age  and  who  shall 
comply  with  the  requirements  of  either  (a)  or  (6)  for  Active  members  and,  in 
addition,  shall  by  his  proficiency  and  contributions  have  attained  to  an  outstand- 
ing rank  among  engineers  or  executives  of  the  motion  picture  industry.  A 
Fellow  shall  be  entitled  to  vote  and  to  hold  any  office  in  the  Society. 

An  Active  member  is  one  who  shall  be  not  less  than  25  years  of  age,  and  shall  be: 

(a)  A  motion  picture  engineer  by  profession.  He  shall  have  been  engaged 
in  the  practice  of  his  profession  for  a  period  of  at  least  three  years,  and  shall  have 
taken  responsibility  for  the  design,  installation,  or  operation  of  systems  or  appa- 
ratus pertaining  to  the  motion  picture  industry. 

(6)  A  person  regularly  employed  in  motion  picture  or  closely  allied  work, 
who  by  his  inventions  or  proficiency  in  motion  picture  science  or  as  an  executive 
of  a  motion  picture  enterprise  of  large  scope,  has  attained  to  a  recognized  standing 
in  the  motion  picture  industry.  In  case  of  such  an  executive,  the  applicant  must 
be  qualified  to  take  full  charge  of  the  broader  features  of  motion  picture  engi- 
neering involved  in  the  work  under  his  direction. 

(c)  An  Active  member  is  privileged  to  vote  and  to  hold  any  office  in  the 
Society. 

An  Associate  member  is  one  who  shall  be  not  less  than  18  years  of  age,  and 
shall  be  a  person  who  is  interested  in  or  connected  with  the  study  of  motion 
picture  technical  problems  or  the  application  of  them.  An  Associate  member  is 
not  privileged  to  vote,  to  hold  office  or  to  act  as  chairman  of  any  committee, 
although  he  may  serve  upon  any  committee  to  which  he  may  be  appointed;  and, 
when  so  appointed,  shall  be  entitled  to  the  full  voting  privileges  of  a  committee 
member. 


346  CONSTITUTION  AND  BY-LAWS  [J.  S.  M.  p.  E. 

A  Sustaining  member  is  an  individual,  a  firm,  or  corporation  contributing 
substantially  to  the  financial  support  of  the  Society. 

Sec.  2. — All  applications  for  membership  or  transfer,  except  for  honorary  or 
fellow  membership,  shall  be  made  on  blank  forms  provided  for  the  purpose,  and 
shall  give  a  complete  record  of  the  applicant's  education  and  experience.  Honor- 
ary and  Fellow  membership  may  not  be  applied  for. 

Sec.  3. — (a)  An  Honorary  membership  may  be  granted  upon  recommendation 
of  the  Board  of  Governors  when  confirmed  by  a  four-fifths  majority  vote  of  the 
Honorary  members,  Fellows,  and  Active  members  present  at  any  regular  meeting 
of  the  Society.  An  Honorary  member  shall  be  exempt  from  all  dues. 

(&)  Fellow  membership  may  be  granted  upon  recommendation  of  at  least 
three-fourths  of  the  Board  of  Governors. 

(c)  Applicants  for  Active  membership  shall  give  as  reference  at  least  three 
members  of  Active  or  of  higher  grade  in  good  standing.     Applicants  shall  be 
elected  to  membership  by  the  approval  of  at  least  three-fourths  of  the  Board  of 
Governors. 

(d)  Applicants  for  Associate  membership  shall  give  as  reference  at  least  one 
member  of  higher  grade  in  good  standing.     Applicants  shall  be  elected  to  member- 
ship by  the  approval  of  at  least  three-fourths  of  the  Board  of  Governors. 

By-Law  II 

Officers 

Sec.  1. — An  officer  or  governor  shall  be  an  Honorary,  a  Fellow,  or  Active  mem- 
ber. 

Sec.  2. — Vacancies  in  the  Board  of  Governors  shall  be  filled  by  the  Board  of 
Governors  until  the  annual  meeting  of  the  Society. 

By-Law  III 

Board  of  Governors 

Sec.  1. — The  Board  of  Governors  shall  transact  the  business  of  the  Society  be- 
tween members'  meetings,  and  shall  meet  at  the  call  of  the  president. 

Sec.  2. — A  majority  of  the  Board  of  Governors  shall  constitute  a  quorum  at 
regular  meetings. 

Sec.  3. — When  voting  by  letter  ballot,  a  majority  affirmative  vote  of  the  total 
membership  of  the  Board  of  Governors  shall  carry  approval,  except  as  otherwise 
provided. 

Sec.  4. — The  Board  of  Governors,  when  making  nominations  to  office,  and 
to  the  Board,  shall  endeavor  to  nominate  persons,  who  in  the  aggregate  are 
representative  of  the  various  branches  or  organizations  of  the  motion  picture  in- 
dustry, to  the  end  that  there  shall  be  no  substantial  predominance  upon  the  Board, 
as  the  result  of  its  own  action,  of  representatives  of  any  one  or  more  branches  or 
organizations  of  the  industry. 

By-Law  IV 

Meetings 

Sec.  1. — The  location  of  each  meeting  of  the  Society  shall  be  determined  by 
the  Board  of  Governors. 


Mar.,  1939]  CONSTITUTION  AND  BY-LAWS  347 

Sec.  2. — Only  Honorary  members,  Fellows,  and  Active  members  shall  be  en- 
titled to  vote. 

Sec.  3. — A  quorum  of  the  Society  shall  consist  in  number  of  one-tenth  of  the 
total  number  of  Honorary  members,  Fellows,  and  Active  members  as  listed  in  the 
Society's  records  at  the  close  of  the  last  fiscal  year. 

Sec.  4. — The  fall  convention  shall  be  the  annual  meeting. 

Sec.  5. — Special  meetings  may  be  called  by  the  president  and  upon  the  request 
of  any  three  members  of  the  Board  of  Governors  not  including  the  president. 

Sec.  6. — All  members  of  the  Society  in  any  grade  shall  have  the  privilege  of  dis- 
cussing technical  material  presented  before  the  Society  or  its  Sections. 

By-Law  V 

Duties  of  Officers 

Sec.  1. — The  President  shall  preside  at  all  business  meetings  of  the  Society  and 
shall  perform  the  duties  pertaining  to  that  office.  As  such  he  shall  be  the  chief 
executive  of  the  Society,  to  whom  all  other  officers  shall  report. 

Sec.  2. — In  the  absence  of  the  president,  the  officer  next  in  order  as  listed  in 
Article  4  of  the  Constitution  shall  preside  at  meetings  and  perform  the  duties  of 
the  president. 

Sec.  3. — The  five  vice-presidents  shall  perform  the  duties  separately  enumerated 
below  for  each  office,  or  as  defined  by  the  president : 

(a)  The  Executive  Vice-President  shall  represent  the  president  in  such  geo- 
graphical areas  of  the  United  States  as  shall  be  determined  by  the  Board  of 
Governors,  and  shall  be  responsible  for  the  supervision  of  the  general  affairs  of 
the  Society  in  such  areas,  as  directed  by  the  president  of  the  Society. 

(b)  The  Engineering  Vice-President  shall  appoint  all  technical  committees. 
He  shall  be  responsible  for  the  general  initiation,  supervision,  and  coordination  of 
the  work  in  and  among  these  committees.     He  may  act  as  chairman  of  any  com- 
mittee or  otherwise  be  a  member  ex-officio. 

(c)  The  Editorial  Vice-President  shall  be  responsible  for  the  publication  of 
the  Society's  JOURNAL  and  all  other  technical  publications.     He  shall  pass  upon 
the  suitability  of  the  material  for  publication,  and  shall  cause  material  suitable 
for  publication  to  be  solicited  as  may  be  needed.     He  shall  appoint  a  papers 
committee  and  an  editorial  committee.     He  may  act  as  chairman  of  any  com- 
mittee or  otherwise  be  a  member  ex-officio. 

(d)  The  Financial  Vice-President  shall  be  responsible  for  the  financial  opera- 
tions of  the  Society,  and  shall  conduct  them  in  accordance  with  budgets  approved 
by  the  Board  of  Governors.     He  shall  study  the  costs  of  operation  and  the  in- 
come possibilities  to  the  end  that  the  greatest  service  may  be  rendered  to  the 
members  of  the  Society  within  the  available  funds.     He  shall  submit  proposed 
budgets  to  the  Board.     He  shall  appoint  at  his  discretion  a  ways  and  means 
committee,  a  membership  committee,  a  commercial  advertising  committee,  and 
such  other  committees  within  the  scope  of  his  work  as  may  be  needed.     He  may 
act  as  chairman  of  any  of  these  committees  or  otherwise  be  a  member  ex-officio. 

(e)  The  Convention  Vice-President  shall  be  responsible  for  the  national  con- 
ventions of  the  Society.     He  shall  appoint  a  convention  arrangements  com- 
mittee, an  apparatus  exhibit  committee,  and  a  publicity  committee.     He  may 
act  as  chairman  of  any  committee,  or  otherwise  be  a  member  ex-officio. 


348  CONSTITUTION  AND  BY-LAWS  [J.  S.  M.  p.  E. 

Sec.  4. — The  Secretary  shall  keep  a  record  of  all  meetings;  he  shall  conduct  the 
correspondence  relating  to  his  office,  and  shall  have  the  care  and  custody  of 
records,  and  the  seal  of  the  Society 

Sec.  5. — The  Treasurer  shall  have  charge  of  the  funds  of  the  Society  and  disburse 
them  as  and  when  authorized  by  the  financial  vice-president.  He  shall  make 
an  annual  report,  duly  audited,  to  the  Society,  and  a  report  at  such  other  times 
as  may  be  requested.  He  shall  be  bonded  in  an  amount  to  be  determined  by  the 
Board  of  Governors  and  his  bond  filed  with  the  secretary. 

Sec.  6. — Each  officer  of  the  Society,  upon  the  expiration  of  his  term  of  office, 
shall  transmit  to  his  successor  a  memorandum  outlining  the  duties  and  policies 
of  his  office. 

By-Law  VI 

Elections 

Sec.  1. — (a)  All  officers  and  five  governors  shall  be  elected  to  their  respective 
offices  by  a  majority  of  ballots  cast  by  the  Active,  Fellow,  and  Honorary  members 
in  the  following  manner : 

Not  less  than  three  months  prior  to  the  annual  fall  convention,  the  Board  of 
Governors,  having  invited  nominations  from  the  Active,  Fellow,  and  Honorary 
membership  by  letter  form  not  less  than  forty  days  before  the  Board  of  Governors' 
meeting,  shall  nominate  for  each  vacancy  several  suitable  candidates.  The  sec- 
retary shall  then  notify  these  candidates  of  their  nomination,  in  order  of  nomina- 
tion, and  request  their  consent  to  run  for  office.  From  the  list  of  acceptances, 
not  more  than  two  names  for  each  vacancy  shall  be  selected  by  the  Board  of 
Governors  and  placed  on  a  letter  ballot.  A  blank  space  shall  also  be  provided 
on  this  letter  ballot  under  each  office,  in  which  space  the  names  of  any  Fellows  or 
Honorary  members  other  than  those  suggested  by  the  Board  of  Governors  may 
be  voted  for.  The  balloting  shall  then  take  place. 

The  ballot  shall  be  enclosed  in  a  blank  envelope  which  is  enclosed  in  an  outer 
envelope  bearing  the  secretary's  address  and  a  space  for  the  member's  name  and 
address.  One  of  these  shall  be  mailed  to  each  Active,  Fellow,  and  Honorary 
member  of  the  Society,  not  less  than  forty  days  in  advance  of  the  annual  fall  con- 
vention. 

The  voter  shall  then  indicate  on  the  ballot  one  choice  for  each  office,  seal  the 
ballot  in  the  blank  envelope,  place  this  in  the  envelope  addressed  to  the  secretary, 
sign  his  name  and  address  on  the  latter,  and  mail  it  in  accordance  with  the  in- 
structions printed  on  the  ballot.  No  marks  of  any  kind  except  those  above  pre- 
scribed shall  be  placed  upon  the  ballots  or  envelopes. 

The  sealed  envelope  shall  be  delivered  by  the  secretary  to  a  committee  of  tell- 
ers appointed  by  the  president  at  the  annual  fall  convention.  This  committee 
shall  then  examine  the  return  envelopes,  open  and  count  the  ballots,  and  announce 
the  results  of  the  election. 

The  newly  elected  officers  and  governors  of  the  general  Society  shall  take  office 
on  the  January  1st  following  their  election. 

(6)  The  first  group  of  vice-presidents,  viz.,  the  executive  vice-president,  engi- 
neering vice-president,  editorial  vice-president,  financial  vice-president,  conven- 
tion vice-president,  and  a  fifth  governor,  shall  be  nominated  by  the  Board  of 
Governors  at  its  first  meeting  after  the  ratification  of  the  corresponding  provisions 


Mar.,  1939]  CONSTITUTION  AND  BY-LAWS  349 

of  the  Constitution;  and  the  membership  shall  vote  on  the  candidates  in  accord- 
ance with  the  procedure  prescribed  in  these  By-Laws  for  regular  elections  of 
officers  so  far  as  these  may  be  applicable. 

By-Law  VII 

Dues  and  Indebtedness 

$ec  i — xhe  annual  dues  shall  be  fifteen  dollars  ($15)  for  Fellows  and  Active 
members  and  seven  dollars  and  fifty  cents  ($7.50)  for  Associate  members,  payable 
on  or  before  January  1st  of  each  year.  Current  or  first  year's  dues  for  new  mem- 
bers, dating  from  the  notification  of  acceptance  in  the  Society,  shall  be  prorated 
on  a  monthly  basis.  Five  dollars  of  these  dues  shall  apply  for  annual  subscription 
to  the  JOURNAL.  No  admission  fee  will  be  required  for  any  grade  of  member- 
ship. 

Sec.  2. — (a)  Transfer  of  membership  may  be  made  effective  at  any  time  by 
payment  of  the  pro  rata  dues  for  the  current  year. 

(b)  No  credit  shall  be  given  for  annual  dues  in  a  membership  transfer  from  a 
higher  to  a  lower  grade,  and  such  transfers  shall  take  place  on  January  1st  of  each 
year. 

(c)  The  Board  of  Governors  upon  their  own  initiative  and  without  a  transfer 
application  may  elect,  by  the  approval  of  at  least  three-fourths  of  the  Board,  any 
Associate  or  Active  member  for  transfer  to  any  higher  grade  of  membership. 

Sec,  5. — Annual  dues  shall  be  paid  in  advance.  All  Honorary  Members,  Fel- 
lows, and  Active  Members  in  good  standing,  as  defined  in  Sec.  5,  may  vote  or 
otherwise  participate  in  the  meetings. 

Sec.  4. — Members  shall  be  considered  delinquent  whose  annual  dues  for  the 
year  remain  unpaid  on  February  1st.  The  first  notice  of  delinquency  shall  be 
mailed  February  1st.  The  second  notice  of  delinquency  shall  be  mailed,  if  neces- 
sary, on  March  1st,  and  shall  include  a  statement  that  the  member's  name  will  be 
removed  from  the  mailing  list  for  the  JOURNAL  and  other  publications  of  the  Society 
before  the  mailing  of  the  April  issue  of  the  JOURNAL.  Members  who  are  in  arrears 
of  dues  on  June  1st,  after  two  notices  of  such  delinquency  have  been  mailed  to 
their  last  address  of  record,  shall  be  notified  their  names  have  been  removed 
from  the  mailing  list  and  shall  be  warned  unless  remittance  is  received  on  or  before 
August  1st,  their  names  shall  be  submitted  to  the  Board  of  Governors  for  action 
at  the  next  meeting.  Back  issues  of  the  JOURNAL  shall  be  sent,  if  available,  to 
members  whose  dues  have  been  paid  prior  to  August  1st. 

Sec.  5. — (a)Members  whose  dues  remain  unpaid  on  October  1st  may  be  dropped 
from  the  rolls  of  the  Society  by  majority  vote  and  action  of  the  Board,  or  the 
Board  may  take  such  action  as  it  sees  fit. 

(6)  Anyone  who  has  been  dropped  from  the  rolls  of  the  Society  for  non-pay- 
ment of  dues  shall,  in  the  event  of  his  application  for  reinstatement,  be  considered 
as  a  new  member. 

(c)  Any  member  may  be  suspended  or  expelled  for  cause  by  a  majority  vote 
of  the  entire  Board  of  Governors ;  provided  he  shall  be  given  notice  and  a  copy 
in  writing  of  the  charges  preferred  against  him,  and  shall  be  afforded  oppor- 
tunity to  be  heard  ten  days  prior  to  such  action. 

Sec.  6. — The  provisions  of  Section  1  to  4,  inclusive,  of  this  By-Law  VII,  given 
above  may  be  modified  or  rescinded  by  action  of  the  Board  of  Governors. 


350  CONSTITUTION  AND  BY-LAWS  [J.  S.  M.  P.  E 

By-Law  VIII 

Emblem 

Sec.  1 . — The  emblem  of  the  Society  shall  be  a  facsimile  of  a  four-hole  film-reel 
with  the  letter  5  in  the  upper  center  opening,  and  the  letters  M,  P,  and  E,  in  the 
three  lower  openings,  respectively.  In  the  printed  emblem,  the  four-hole  open- 
ings shall  be  orange,  and  the  letters  black,  the  remainder  of  the  insignia  being  black 
and  white.  The  Society's  emblem  may  be  worn  by  members  only. 

By-Law  IX 

Publications 

Sec.  1 . — Papers  read  at  meetings  or  submitted  at  other  times,  and  all  material 
of  general  interest  shall  be  submitted  to  the  editorial  board,  and  those  deemed 
worthy  of  permanent  record  shall  be  printed  in  the  JOURNAL.  A  copy  of  each 
issue  shall  be  mailed  to  each  member  in  good  standing  to  his  last  address  of  record. 
Extra  copies  of  the  JOURNAL  shall  be  printed  for  general  distribution  and  may  be 
obtained  from  the  General  Office  on  payment  of  a  fee  fixed  by  the  Board  of 
Governors. 

By-Law  X 

Local  Sections 

Sec.  1. — Sections  of  the  Society  may  be  authorized  in  any  state  or  locality  where 
the  Active,  Fellow,  and  Honorary  membership  exceeds  20.  The  geographic 
boundaries  of  each  Section  shall  be  determined  by  the  Board  of  Governors. 

Upon  written  petition,  signed  by  20  or  more  Active  members,  Fellows  and 
Honorary  members,  for  the  authorization  of  a  Section  of  the  Society,  the  Board  of 
Governors  may  grant  such  authorization. 

MEMBERSHIP 

Sec.  2. — All  members  of  the  Society  of  Motion  Picture  Engineers  in  good  stand- 
ing residing  in  that  portion  of  any  country  set  apart  by  the  Board  of  Governors 
tributary  to  any  local  Section  shall  be  eligible  for  membership  in  that  Section,  and 
when  so  enrolled  they  shall  be  entitled  to  all  privileges  that  such  local  Section 
may,  under  the  General  Society's  Constitution  and  By-Laws,  provide. 

Any  member  of  the  Society  in  good  standing  shall  be  eligible  for  non-resident 
affiliated  membership  of  any  Section  under  conditions  and  obligations  prescribed 
for  the  Section.  An  affiliated  member  shall  receive  all  notices  and  publications 
of  the  Section  but  he  shall  not  be  entitled  to  vote  at  Sectional  meetings. 

Sec.  3. — Should  the  enrolled  Active,  Fellow,  and  Honorary  membership  of  a 
Section  fall  below  20,  or  should  the  technical  quality  of  the  presented  papers  fall 
below  an  acceptable  level,  or  the  average  attendance  at  meetings  not  warrant  the 
expense  of  maintaining  the  organization,  the  Board  of  Governors  may  cancel  its 
authorization. 

OFFICERS 

Sec.  4. — Each  Section  shall  nominate  and  elect  a  chairman,  two  managers,  and 
a  secretary-treasurer.  The  Section  chairmen  shall  automatically  become  mem- 
bers of  the  Board  of  Governors  of  the  General  Society,  and  continue  in  that  posi- 
tion for  the  duration  of  their  terms  as  chairmen  of  the  local  Sections. 


Mar.,  1939]  CONSTITUTION  AND  BY-LAWS  351 

ELECTION  OF  OFFICERS 

Sec.  5. — The  officers  of  a  Section  shall  be  Active,  Fellow,  or  Honorary  members 
of  the  General  Society.  They  shall  be  nominated  and  elected  to  sectional  office 
under  the  method  prescribed  under  By-Law  VI,  Section  1,  for  the  nomination 
and  election  of  officers  of  the  General  Society.  The  word  manager  shall  be  sub- 
stituted for  the  word  governor.  All  Section  officers  shall  hold  office  for  one  year, 
or  until  their  successors  are  chosen,  except  the  Board  of  Managers,  as  hereinafter 
provided. 

MANAGERS 

$ec  Q — The  Board  of  Managers  shall  consist  of  the  Section  chairman,  the  Sec- 
tion past-chairman,  the  Section  secretary-treasurer,  and  two  Active,  Fellow,  or 
Honorary  members,  one  of  which  last  named  shall  be  elected  for  a  two-year  term, 
and  one  for  one  year,  and  then  one  for  two  years  each  year  thereafter.  At  the 
discretion  of  the  Board  of  Governors,  and  with  their  written  approval,  this  list 
of  officers  may  be  extended. 

BUSINESS 

Sec.  7. — The  business  of  a  Section  shall  be  conducted  by  the  Board  of  Managers. 

EXPENSES 

Sec.  8. — (a)  As  early  as  possible  in  the  fiscal  year,  the  secretary  of  each  Section 
shall  submit  to  the  Board  of  Governors  of  the  Society  a  budget  of  expenses  for  the 
year. 

(b)  The  treasurer  of  the  General  Society  may  deposit  with  each  Section  secre- 
tary-treasurer a  sum  of  money,  the  amount  to  be  fixed  by  the  Board  of  Governors, 
for  current  expenses. 

(c)  The  secretary-treasurer  of  each  Section  shall  send  to  the  treasurer  of  the 
General  Society,  quarterly  or  on  demand,  an  itemized  account  of  all  expenditures 
incurred  during  the  preceding  interval. 

(d)  Expenses  other  than  those  enumerated  in  the  budget,  as  approved  by  the 
Board  of  Governors  of  the  General  Society,  shall  not  be  payable  from  the  general 
funds  of  the  Society  without  express  permission  from  the  Board  of  Governors. 

(e)  A  Section  Board  of  Managers  shall  defray  all  expenses  of  the  Section  not 
provided  for  by  the  Board  of  Governors,  from  funds  raised  locally  by  donation,  or 
by  fixed  annual  dues,  or  by  both. 

(/)  The  secretary  of  the  Society  shall,  unless  otherwise  arranged,  supply  to 
each  Section  all  stationery  and  printing  necessary  for  the  conduct  of  its  business. 

MEETINGS 

Sec.  9. — The  regular  meetings  of  a  Section  shall  be  held  in  such  places  and  at 
such  hours  as  the  Board  of  Managers  may  designate. 

The  secretary-treasurer  of  each  Section  shall  forward  to  the  secretary  of  the 
General  Society,  not  later  than  five  days  after  a  meeting  of  a  Section,  a  statement 
of  the  attendance  and  of  the  business  transacted. 


352  CONSTITUTION  AND  BY-LAWS 

PAPERS 

Sec.  10. — Papers  shall  be  approved  by  the  Section's  papers  committee  previ- 
ously to  their  being  presented  before  a  Section.  Manuscripts  of  papers  presented 
before  a  Section,  together  with  a  report  of  the  discussions  and  the  proceedings  of 
the  Section  meetings,  shall  be  forwarded  promptly  by  the  Section  secretary- 
treasurer  to  the  secretary  of  the  General  Society.  Such  material  may,  at  the  dis- 
cretion of  the  board  of  editors  of  the  General  Society,  be  printed  in  the  Society's 
publications. 

CONSTITUTION  AND  BY-LAWS 

Sec.  11. — Sections  shall  abide  by  the  Constitution  and  By-Laws  of  the  Society, 
and  conform  to  the  regulations  of  the  Board  of  Governors.  The  conduct  of  Sec- 
tions shall  always  be  in  conformity  with  the  general  policy  of  the  Society  as  fixed 
by  the  Board  of  Governors. 

By-Law  XI 

Amendments 

Sec.  1. — These  By-Laws  may  be  amended  at  any  regular  meeting  of  the  So- 
ciety by  the  affirmative  vote  of  two-thirds  of  the  members  present  at  a  meeting 
who  are  eligible  to  vote  thereon,  a  quorum  being  present,  either  on  the  recom- 
mendation of  the  Board  of  Governors  or  by  a  recommendation  to  the  Board 
of  Governors  signed  by  any  ten  members  of  active  or  higher  grade,  provided  that 
the  proposed  amendment  or  amendments  shall  have  been  published  in  the  JOUR- 
NAL of  the  Society,  in  the  issue  next  preceding  the  date  of  the  stated  business  meet- 
ing of  the  Society,  at  which  the  amendment  or  amendments  are  to  be  acted  upon. 

Sec.  2. — In  the  event  that  no  quorum  of  the  voting  members  is  present  at  the 
time  of  the  meeting  referred  to  in  Sec.  1,  the  amendment  or  amendments  shall  be 
referred  for  action  to  the  Board  of  Governors.  The  proposed  amendment  or 
amendments  then  become  a  part  of  the  By-Laws  upon  receiving  the  affirmative 
vote  of  three-quarters  of  the  Board  of  Governors. 


JOURNAL 

OF  THE  SOCIETY  OF 

MOTION  PICTURE  ENGINEERS 


Volume  XXXII  April,  1939 


CONTENTS 

Page 

A  Motion  Picture  Dubbing  and  Scoring  Stage 

C.  L.  LOOTENS,  D.  J.  BLOOMBERG,  AND  M.  RETTINGER    357 

Unidirectional  Microphone  Technic 

J.  P.  LlVADARY  AND  M.  RETTINGER      381 

Artificially  Controlled  Reverberation S.  K.  WOLF    390 

The  Evolution  of  Arc  Broadside  Lighting  Equipment 

PETER  MOLE     398 

Some  Studies  on  the  Use  of  Color  Coupling  Developers  for 
Toning  Processes K.  FAMULENER    412 

The  Metro-Goldwyn-Mayer  Semi-Automatic  Follow-Focus  De- 
vice  J.  ARNOLD    419 

Independent  Camera   Drive  for   the  A-C.    Interlock   Motor 
System F.  G.  ALBIN    424 

Silent  Wind-Machine F.  G.  ALBIN     430 

New  Motion  Picture  Apparatus 

Characteristics  of  Supreme  Panchromatic  Negative 

A.  W.  COOK    436 
New  Background  Projector  for  Process  Cinematography.  . .  . 

G.  H.  WORRALL    442 

Book  Reviews 445 

Current  Literature 447 

Officers  and  Governors  of  the  Society 452 

Committees  of  the  Society 455 

Spring  Convention  at  Hollywood,  April  17-21,  1939 460 

Abstracts  of  Convention  Papers : 465 

Society  Announcements 480 


JOURNAL 

OF  THE  SOCIETY  OF 

MOTION  PICTURE  ENGINEERS 


SYLVAN  HARRIS,  EDITOR 

Board  of  Editors 
J.  I.  CRABTREE,  Chairman 

A.  N.  GOLDSMITH  A.  C.  HARDY  H.  G.  KNOX 

J.  G.  FRAYNE  L.  A.  JONES  G.  E.  MATTHEWS 

E.  W.  KELLOGG 


Subscription  to  non-members,  $8.00  per  annum;  to  members,  $5.00  per  annum, 
included  in  their  annual  membership  dues;  single  copies,  $1.00.  A  discount 
on  subscription  or  single  copies  of  15  per  cent  is  allowed  to  accredited  agencies. 
Order  from  the  Society  of  Motion  Picture  Engineers,  Inc.,  20th  and  Northampton 
Sts.,  Easton,  Pa.,  or  Hotel  Pennsylvania,  New  York,  N.  Y. 
Published  monthly  at  Easton,  Pa.,  by  the  Society  of  Motion  Picture  Engineers. 

Publication  Office,  20th  &  Northampton  Sts.,  Easton,  Pa. 
General  and  Editorial  Office,  Hotel  Pennsylvania,  New  York,  N.  Y. 

West-Coast  Office,  Suite  226,  Equitable  Bldg.,  Hollywood,  Calif. 
Entered  as  second  class  matter  January  15,  1930,  at  the  Post  Office  at  Easton, 
Pa.,  under  the  Act  of  March  3,  1879.     Copyrighted,  1939,  by  the  Society  of 
Motion  Picture  Engineers,  Inc. 

Papers  appearing  in  this  Journal  may  be  reprinted,  abstracted,  or  abridged 
provided  credit  is  given  to  the  Journal  of  the  Society  of  Motion  Picture  Engineers 
and  to  the  author,  or  authors,  of  the  papers  in  question.  Exact  reference  as  to 
the  volume,  number,  and  page  of  the  Journal  must  be  given.  The  Society  is  not 
responsible  for  statements  made  by  authors. 


OFFICERS  OF  THE  SOCIETY 

**  President:    E.  A.  WILLIFORD,  30  East  42nd  St.,  New  York,  N.  Y. 
**  Past-President:    S.  K.  WOLF,  RKO  Building,  New  York,  N.  Y. 
**  Executive  Vice-President:    N.  Levinson,  Burbank,  Calif. 

*  Engineering  Vice-President:    L.  A.  JONES,  Kodak  Park,  Rochester,  N.  Y. 
**  Editorial  Vice-President:    J.  I.  CRABTREE,  Kodak  Park,  Rochester,  N.  Y. 

*  Financial  Vice-President:    A.  S.  DICKINSON,  28  W.  44th  St.,  New  York,  N.  Y. 
**  Convention  Vice-President:    W.  C.  Kunzmann,  Box  6087,  Cleveland,  Ohio. 

*  Secretary:    J.  FRANK,  JR.,  90  Gold  St.,  New  York,  N.  Y. 

*  Treasurer:    L.  W.  DAVEE,  153  Westervelt  Ave.,  Tenafly,  N.  J. 

GOVERNORS 
**  M.  C.  BATSEL,  Front  and  Market  Sts.,  Camden,  N.  J. 

*  R.  E.  FARNHAM,  Nela  Park,  Cleveland,  Ohio. 

*  H.  GRIFFIN,  90  Gold  St.,  New  York,  N.  Y. 

*  D.  E.  HYNDMAN,  350  Madison  Ave.,  New  York,  N.  Y. 

*  L.  L.  RYDER,  5451  Marathon  St.,  Hollywood,  Calif. 

*  A.  C.  HARDY,  Massachusetts  Institute  of  Technology.  Cambridge,  Mass. 

*  S.  A.  LUKES,  6427  Sheridan  Rd.,  Chicago,  111. 

**  H.  G.  TASKER,  14065  Valley  Vista  Blvd.,  Van  Nuys,  Calif. 

*  Term  expires  December  31,  1939. 
**  Term  expires  December  31,  1940. 


A  MOTION  PICTURE  DUBBING  AND  SCORING  STAGE* 

C.  L.  LOOTENS,**  D.  J.  BLOOMBERG,**  AND  M.  RETTINGERt 

Summary. — A  new  dubbing  (re-recording)  and  scoring  (music  recording)  building 
recently  completed  on  the  Republic  lot  consists  of  a  recording  stage,  scoring  monitor 
room,  machine  room,  projection  booth,  power  room,  maintenance  room,  and  a  recording 
truck  testing  platform. 

The  recording  equipment  consists  essentially  of  two  complete  RCA  high-fidelity 
recording  channels,  together  with  their  associated  equipment  of  film  recorders,  film 
phonographs,  amplifier  racks,  power  rectifiers,  dubbing  and  scoring  consoles,  "ace- 
tate" recorder  and  projection  equipment. 

The  stage  is  of  the  live-end-dead-end  type,  and  has  dimensions  which  conform  to 
the  recommended  1:2:3  ratio.  The  live-end  is  provided  with  permanent  side  wall  and 
ceiling  reflecting  panels  which  increase  the  reverberation  and  diffusion  of  sound.  The 
remainder  of  the  stage  is  treated  with  4-inch  rockwool  battens,  placed  between  the  studs 
and  retained  in  place  by  a  dual  muslin  covering.  The  measured  reverberation  charac- 
teristic of  the  stage,  fulfilling  recommended  requirements,  is  between  0.95  and  1 .00 
second  for  the  frequency  band  of  540  to  7000  cps.  The  stage  is  equipped  also  with  an 
eight-position  mixer  console  so  that  dubbing  may  be  monitored  in  a  room  having  ap- 
proximate theater  sound  characteristics. 

Republic's  recently  completed  dubbing  (re-recording)  and  scoring 
(music  recording)  building  was  designed  and  constructed  essentially 
for  sound  recording.  The  building  is  located  adjacent  to  the  sound- 
cutting  building  and  readily  accessible  to  the  film  vaults  and  loading 
rooms.  The  complete  unit  (Fig.  1)  consists  of  a  recording  stage, 
scoring  monitor  room,  machine  room,  projection  booth,  power  room, 
maintenance  room,  and  recording-truck  testing  platform. 

When  designing  the  building,  acoustical  considerations  and  acces- 
sibility between  rooms  were  given  preference.  Reinforced  concrete 
footings  and  foundations  are  used  for  all  walls  and  supporting  columns 
making  the  building  rigid  and  free  from  vibration.  The  stage  foot- 
ings and  foundation  are  isolated  from  those  of  the  remainder  of  the 
building  by  a  6-inch  space  filled  with  granulated  cork.  All  ground 

*  Presented  at  the  1938  Fall  Meeting  at  Detroit,  Mich. ;  received  October  27, 
1938. 

'**  Republic  Productions,  Inc.,  North  Hollywood,  Calif. 
t    RCA  Manufacturing  Co.,  Hollywood,  Calif. 

357 


358 


LOOTENS,  BLOOMBERG,  AND  RETTINGER    [J.  S.  M.  p.  E. 


floors  are  constructed  of  fine  gravel  and  sand  asphalt  laid  hot  within 
the  foundation  walls.  Asphalt  was  used  in  preference  to  concrete  be- 
cause the  asphalt  has  a  lower  sound- transmission  rate  and  is  compara- 
tively non-homogeneous.  The  walls  and  roof  structure  of  the  work- 
rooms adjacent  to  the  stage  are  of  wood  construction.  The  outsides 

of  the  walls  are  covered  with  1-inch 
diagonal  sheathing  and  a  1-inch  finish 
of  exterior  cement  stucco.  The  in- 
terior walls  are  of  plaster  board  and 
cement  plaster  finish.  The  ceilings  of 
all  rooms  and  the  hallway,  with  the 
exception  of  the  truck- testing  plat- 
form, are  of  plaster  board — y2-inch 
brown  plaster  finished  with  1/2-inch 
"Acoustite"  rockwool  composition 
plaster. 

Recording  Stage. — The  recording 
stage,  which  was  constructed  essen- 
tially as  a  unit,  is  25  feet  high,  50  feet 
wide,  and  75  feet  long,  satisfying  the 
well  known  criterion  that,  for  a  stage 
of  this  size,  the  ratio  of  height,  width, 
and  length  should  be  1 : 2 : 3 .  The  ratio 
of  2:3:5  for  the  above  dimensions  is 
sometimes  recommended,  but  a  little 
consideration  will  quickly  show  that 
an  enclosure  of  such  size  will  not  only 
make  an  undesirably  high  ceiling  and 
narrowly  spaced  side  walls,  but  will 
also  cause  the  "mean  free  path" 
(average  length  of  one  reflection)  to 
be  longer.  This  condition  produces 
fewer  reflections  per  second  at  any 
point  in  the  room,  resulting  in  de- 
creased diffusion  of  sound  in  the  room. 


FIG.  1.     Plan  of  buildings. 


Forty  musicians  represent  the  optimum  number  of  performers  in 
this  stage  for  maximum  quality  and  best  illusion,  while  eighty  musi- 
cians represent  the  largest  number  that  can  be  crowded  into  the  space 
and  still  provide  reasonably  satisfactory  recordings.  In  this  connec- 
tion it  should  be  remembered  that  it  is  practically  impossible  to  imi- 


April,  1939] 


DUBBING  AND  SCORING  STAGE 


359 


tate  the  acoustics  of  a  large  room  in  an  enclosure  that  is  actually 
small.  On  the  other  hand,  the  acoustics  of  a  small  room  can  be  imi- 
tated easily  in  a  large  enclosure  by  the  judicious  use  of  flats  and  other 
reflecting  surfaces. 

Only  two  materials,  rockwool  and  wood,  are  used  for  the  interior 
finish  of  this  live-end-dead-end  stage.  In  Figs.  1,  2,  and  3,  it  is  seen 
that  wood  is  used  as  a  reflective  material  around  the  band-shell  (live- 
end)  to  produce  sufficient  localized  reverberation  to  permit  the  musi- 
cians, long  accustomed  to  playing  in  reverberant  concert  halls,  to  keep 
more  easily  in  tune,  to  determine  without  undue  effort  precisely  the 
true  pitch  of  the  following  note  while  perceiving  the  present  one,  and 
to  retain  proper  balance  between  bass  and  treble.  Wood  instead  of 


Alltlta"  Book  Wool  Treatment 


FIG.  2.     Elevation  of  stage,  monitor  room,  and  projection  room. 

some  other  material  such  as  plaster  or  hard  pressed-board  was  se- 
lected because  of  its  well  liked  and  practically  inimitable  quality  of 
resonating  over  a  wide  range  of  musical  pitch,  as  evidenced  by  its  use 
in  concert  halls  of  avowedly  superior  acoustics. 

Because  of  the  reverberant  character  of  this  band-shell  it  became 
important  to  make  arrangements  for  the  prevention  of  echoes  and  for 
an  effective  dispersion  of  the  sound  to  obtain  uniform  distribution  of  it 
in  the  stage.  This  object  was  achieved  by  providing  suitably  orien- 
tated corrugations  of  sufficient  depth  to  become  effective  also  as  dif- 
fusers  of  sound  for  lower  frequencies.  The  absorbent  panels  at  re- 
versed angles  to  the  reflective  ones  are  necessary  to  prevent  the  sound 
from  being  returned  into  the  shell  in  too  great  a  measure. 

The  orientation  of  the  reflective  side-wall  and  ceiling  splays  is  such 
as  to  obtain  not  only  sufficient  localized  reverberation  in  the  band- 
shell  but  also  to  achieve  a  desirably  directed  efflux  of  the  sound  into 
the  dead  end  of  the  stage.  Reflecting  surfaces  close  to  the  perform- 
ers are,  therefore,  positioned  so  as  to  secure  enough  short- time  reflec- 


360 


LOOTENS,  BLOOMBERG,  AND  RETTINGER    [j.  s.  M.  p.  E. 


tions  at  the  microphone  to  preserve  the  naturalness  of  the  instru- 
ments, while  more  distant  reflecting  surfaces  are  utilized  to  produce 
in  the  band-shell  a  sufficient  amount  of  the  long-time  reflections  which 
musically  are  so  pleasing  and  without  which  the  music  would  tend  to 
sound  flat,  as  in  the  open  air.  The  problem,  on  close  inspection,  will 
be  found  to  be  exceedingly  complex,  not  only  because  it  is  essentially 
three-dimensional  in  character,  but  also  because  not  one  but  many 
sources  of  sound  must  be  taken  into  consideration.  A  solution  was 
obtained  by  drawing  many  diagrams  depicting  the  travel  of  sound 


FIG.  3.     Interior  of  recording  stage. 

from  various  parts  in  the  stage  and  for  several  angles  of  splay-slope, 
and  then  choosing  a  splay  orientation  that  gave  the  most  desirable 
results  for  the  largest  number  of  considered  imaginary  point-sources 
representing  the  musicians. 

The  absorbent  regions  surrounding  the  microphone,  as  well  as  the 
dead  end  of  the  stage  are,  of  course,  necessary  to  provide  placement  of 
sufficient  damping  material  to  give  the  desired  reverberation  period 
in  the  stage  and  to  permit  the  smoothing  out  of  whatever  interference 
pattern  may  exist  there.  Interference,  it  is  well  known,  may  be  of 
the  nature  of  a  space  or  a  time-effect.  The  first  enters  most  clearly 
during  sustained  passages,  when  the  transmitter  may  be  at  a  region  of 


April,  1939] 


DUBBING  AND  SCORING  STAGE 


361 


reduced  or  enhanced  sound-intensity.  The  time-effect  makes  itself 
known  in  the  form  of  irregular  fluctuations  of  the  sound-pressure  dur- 
ing growth  or  decay  of  the  sound  in  the  room.  The  more  reverberant 
a  room,  the  more  pronounced  are  the  two  effects,  and  the  more  dis- 
turbing do  they  appear  in  a  recording. 


*"    .6 

5 

i: 


VOLUME   IK  COB»C  FEET 


FIG. 


\OO  1.000  10.000 

FREQUENCY  -*• 

FIG.  4.     ( Upper)     Reverberation-volume  curve. 
5.     (Lower)    Republic    stage    reverberation-frequency 
curve. 


Recording  studios  should  have  a  reverberation- time  at  1000  cycles 
of  about  two-thirds  of  that  found  satisfactory  for  a  room  of  equal  vol- 
ume used  for  binaural  hearing.  Fig.  4  shows  the  limits  of  variation 
of  reverberation  time  with  volume  for  scoring  stages,  and  illustrates 
that  it  is  not  readily  possible  to  speak  of  the  optimal  time  of  rever- 
beration of  a  room  as  a  definite  figure  unless  one  mentions  at  the  same 
time  the  type  of  activity  for  which  the  room  is  being  used,  such  as  for 
speech,  piano  recitals,  or  songs.  The  optimal  time,  in  general,  is 
therefore  a  range,  with  the  upper  limit  pertaining  to  organ  oratoria 


362  LOOTENS,  BLOOMBERG,  AND  RETTINGER    [J.  S.  M.  P.  E. 

and  the  lower  limit  to  speech.  A  mean  value  of  reverberation  at 
1000  cycles  of  0.97  second  was  chosen  as  being  the  time  most  suitable 
for  the  type  of  music  to  be  most  frequently  played  on  this  stage.  A 
number  of  portable  hinged  panels,  4  by  7  feet,  absorbent  on  one  side 
and  reflective  on  the  other,  are  used  to  provide  acoustical  isolation  or 
special  reverberation  effects  for  small  groups  of  performers. 

The  variation  of  reverberation  time  with  frequency,  called  the 
reverberation  characteristic,  has  been  widely  discussed  in  the  litera- 
ture. 1'2-3-4  However,  none  of  the  criteria  so  far  proposed  for  the  re- 
verberation characteristic  in  rooms  appears  to  be  completely  tenable 
for  scoring  stages.  Even  MacNair's  criterion,*  which  for  the  lower 
frequencies  makes  for  reverberation-times  shorter  than  those  ob- 
tained by  any  other  criterion,  when  applied  to  a  scoring  stage  still 
produces  recordings  marred  by  a  little  "boominess."  Fig.  5  shows  the 
reverberation  characteristic  of  this  scoring  stage,  and  all  tests  made 
so  far — direct  listening  tests  as  well  as  recording  tests — show  that  the 
studio  is  remarkably  free  from  any  boominess  or  prolonged  reverbera- 
tion for  the  lower  frequencies. 

It  may  be  of  interest  to  determine  for  this  stage  the  value  of  the 
M.  O.  Strutt  equation  for  the  ratio  of  the  "useful"  to  delayed  sound. 
By  "useful"  sound  is  meant  the  direct  sound  plus  that  sound  that 
comes  to  an  auditor  within  Vie  second  after  it  was  emitted;  by  de- 
layed sound  is  meant  the  sound  that  comes  to  an  auditor  Vie  second 
after  it  was  emitted.  The  value  of  this  equation  depends,  of  course, 
on  the  distance  between  the  auditor  and  the  source  of  sound ;  for  this 
purpose  an  average  distance  D  equal  to  V1  /3/2  ( V  =  volume  of  stage) 
was  chosen.  The  M.  O.  Strutt5  equation  is: 

+  £     /   e-13.8t/Tdt 


-i3.8//r  dt 

where  P    =  power  output  of  source  of  sound. 
c     =  velocity  of  sound. 

T  =  time  of  reverberation  (the  reverberation  time  for  a  given  frequency 
is  the  time  required  for  the  average  sound-energy  density,  initially  in 
a  steady  state,  to  decrease,  after  the  source  is  stopped,  to  one-mil- 
lionth of  its  initial  value). 

*  By  this  criterion  the  loudness  level  of  all  frequency  components  in  speech 
and  music  should  decay  at  the  same  constant  rate. 


April,  1939]  DUBBING  AND  SCORING  STAGE  363 

V    =  volume  of  room. 
t    =  time. 

D    =  distance  between  source  of  sound  and  position  of  listener  or  micro- 
phone. 

This  equation  may  be  written  as : 

Q  -0.86  [-0.004  y/»  -nix 
eT    L         r  J 

if  r  =  1  second,  V  =  90,000  cubic-feet;  Q  equals  1.78,  showing  that 
the  amount  of  "useful"  sound  is  quite  large  at  this  average  distance 
of  90,000' /3/2  or  22.5  ft. 

All  construction  in  the  stage  was  made  exceedingly  rigid  to  avoid 
structural  resonances.  The  wall-studs  are  2  by  6  inch,  covered  on  the 
outside  with  1-inch  diagonal  sheathing  and  1-inch  exterior  cement 
stucco.  The  entire  wall  area  in  the  dead-end  half  is  treated  with  4- 
inch  long-fiber  rockwool  battens  placed  between  the  studs  and  held 
in  place  by  a  retaining  layer  of  40-44  muslin  stapled  to  the  interior 
side  of  the  studs  (Fig.  8) .  On  top  of  this  layer  of  muslin,  1  by  2-inch 
wood  retaining  strips  are  nailed  at  right  angles  to  the  studs.  The 
strips  are  spaced  on  9-inch  centers  from  the  floor  to  a  5-ft.  height. 
From  this  point  the  spacing  is  graduated  from  18-inch  centers  to  12- 
inch  centers  at  the  plate  line.  On  top  of  the  wood  strips,  a  layer  of 
fire-proofed  and  color-dyed  muslin  is  tacked.  The  tacks  and  muslin 
joints  are  covered  with  3/4-inch  wood  half-round.  Fig.  3  shows  the 
finished  appearance  of  the  absorbent  treatment. 

The  roof  of  the  stage  consists  of  a  Lamella  roof  resting  on  8  by  10-in. 
sills  anchored  to  the  2  by  6-in.  side  walls.  The  crown  of  the  roof  is 
approximately  34x/2  feet  above  the  finished  floor.  The  roof  sills  are 
tied  together  by  four  \l/z-in.  steel  tie-rods  to  absorb  the  roof  thrust. 
The  Lamellas  consist  of  segments  of  an  arc  milled  so  as  to  intersect 
continuously  and  uniformly  as  the  "skew  arch"  crosses  the  roof. 

These  intersecting  arches  (Fig.  6)  form  the  familiar  diamond  pat- 
tern of  the  Lamella  roof.  Each  diamond  in  the  stage  is  8  feet  long  and 
36  inches  wide,  with  each  intersection  joint  bolted  with  one  5/g-inch 
bolt.  The  diamond-patterned  Lamella  roof  is  used  to  carry  the  center 
43-ft.  section  of  the  roof  with  diagonal  sheathed  rafters  closing  the  16- 
foot  end-sections,  and  spanned  from  the  end-walls  to  the  end  arch  of 
the  Lamella  web. 

The  height  of  the  roof  from  the  floor,  the  high  degree  of  fire  resis- 
tance, the  absence  of  intermediate  supports  for  the  roof,  and  the  free- 
dom from  large  structural  members  requiring  sound-deadening  treat- 
ment were  all  factors  guiding  the  selection  of  the  roof  as  the  best 


364 


LOOTENS,  BLOOMBERG,  AND  RETTINGER    [J.  S.  M.  P.  E. 


suited  for  this  modern  recording  stage.  The  entire  roof  area  is  acous- 
tically treated  with  4-inch  rockwool  battens  (Fig.  6)  in  the  same  man- 
ner as  the  wall  area  with  the  exception  of  the  spacing  of  the  retaining 
strips,  which  is  graduated  from  26  inches  at  the  sill  line  to  8  inches  at 
the  center  of  the  ceiling. 

The  interior  of  the  live-end  wall  area  of  the  stage  is  covered  with  1- 
inch  fiber  insulation  board.     The  reflective  side-wall  splays  (Figs.  1, 


FIG.  6.     Roof  construction  and  treatment. 


2,  3)  reaching  from  floor  to  ceiling  consist  of  alternate  wood  panels 
parallel  to  each  other  and  separated  by  angular  absorbing  panels  2 
feet  in  width.  The  panel  construction  consists  of  2  by  6-inch  studs 
and  cross-pieces,  spaced  on  2-ft.  centers  covered  with  1-inch  diagonal 
sheathing  surfaced  with  3/Vinch  grooved  T  &  G  ceiling  lumber.  The 
rear  of  the  panels  is  braced  at  6V2-ft.  intervals  to  the  wall  of  the  stage. 
Eight  35-ft.  angular  ceiling  splays  (Fig.  3)  are  anchored  to  the  ceiling 
between  the  side-wall  panels.  These  panels  are  of  1-inch  plywood 
lumber,  screwed  to  2  by  12-in.  supporting  joists  and  cross-pieces 


April,  1939] 


DUBBING  AND  SCORING  STAGE 


365 


spaced  on  2-f t.  centers  and  bolted  to  the  roof  rafters.  The  surfaces  of 
all  wood  panels  are  stained  and  finished  with  three  coats  of  trans- 
parent varnish. 

The  stage  floor  consists  of  2  by  6-in.  joists  laid  flat  on  the  4-inch 
asphalt  ground  floor.  The  space  between  the  joists  is  filled  to  the 
surface  with  fine  aggregate  asphalt.  A  subfloor  of  1-inch  T  &  G  di- 
agonal sheathing  is  securely  nailed  to  the  joists.  A  1-inch  T  &  G- 
finished  floor  is  nailed  directly  on  top  of  the  sheathing.  This  con- 
struction results  in  a  very  rigid  non-resonant  type  of  floor. 


CANVAS   COVERED 

ROBBER    TUBING  STOP 


I     PLYWOOD 


1 1  AE    NAILING    STRIP 


CANVAS  COVERED 
RUBBER  TUBING  STOI 

I"X3"   OAK 

18  GA.    CALV.  MS.TAL. 
FIBER    BOARD 

l"  PLYWOOD 


FIG.  7.     Detail  of  sound  retarding  doors. 


A  horn-tower  (Figs.  1,  2)  6  feet  wide  by  9  feet  long  is  provided  in 
back  of  the  projection  screen.  The  interior  walls  and  ceiling  of  the 
horn-tower  are  treated  with  4-inch  rockwool  battens.  The  horn- 
tower  floor,  which  is  6  feet  above  the  stage  floor,  is  of  wood  construc- 
tion covered  with  1-inch  fiber  insulation  board.  Entrance  to  the 
tower  is  through  a  trap-door  in  the  floor.  The  sound-reproducing 
system  is  the  well  known  RCA  two-way  loud  speaker  system  em- 
ploying two  high-frequency  and  four  low-frequency  units.  The 
width  of  the  horn-tower  allows  only  a  4-inch  space  between  the  low- 
frequency  units  and  the  side  wall.  This  space  and  the  area  around 
the  high-frequency  speaker  is  closed  in  with  1-in.  fiber  insulation 
board.  This  construction  prevents  the  generation  of  backstage  reso- 
nance in  the  loud  speaker  cavity. 


366  LOOTENS,  BLOOMBERG,  AND  RETTINGER    [j.  s.  M.  p.  E. 

All  openings  leading  into  the  stage  are  closed  with  massive  double 
doors  made  of  dual  sections  of  1-inch  plywood,  1-inch  insulation 
board,  and  18-gauge  sheet-metal,  as  shown  in  Fig.  7.  The  doors  are 
held  tightly  closed  against  rubber  weather-stripping  by  special  non- 
slipping  clamps. 

Recording  wiring  for  scoring  activities  terminates  on  the  stage  in 
conveniently  located  panels  consisting  of  a  6-position  microphone 
outlet  panel,  head-phone  outlet  panel,  signal-light  panel,  and  mobile 
acetate  recorder-outlet  panel.  The  music  director's  stand  is  equipped 
with  a  built-in  PA  microphone,  speaker,  and  head-phone  volume  con- 
trol. A  general  PA  speaker  is  mounted  on  the  rear  wall  of  the  stage. 


FIG.  8.     Republic  dubbing  console. 

The  stage  is  used  also  as  a  monitoring  room  for  dubbing.  For  this 
purpose  an  8-position  mixer  console  is  located  in  the  dead  end  of  the 
stage,  as  shown  in  Figs.  1  and  3.  The  dubbing  console  (Fig.  8)  will 
seat  three  mixer  operators,  and  is  designed  with  a  group  of  special 
equalizer  controls  located  in  the  center  of  the  console.  High-,  low-,  and 
middle-frequency  attenuation  and  equalization,  with  the  various 
telephone,  PA,  and  special  equalizers  are  provided.  Eight  single- 
stage  booster  amplifiers,  used  with  the  equalizers,  are  mounted  in  the 
console,  and  are  accessible  through  panels  in  the  side  or  rear.  A 
variable  high-pass  filter  is  included  with  five  cut-off  frequencies  be- 
tween 80  and  150  cycles.  Telephone,  PA,  and  signal  facilities  con- 
necting with  all  stations,  are  located  within  easy  reach  of  all  the 
mixers. 


April,  1939] 


DUBBING  AND  SCORING  STAGE 


367 


Jack  bays  provide  connections  to  the  machine  room  amplifier  bays, 
scoring  console,  and  for  patching  a  reproducer  through  any  desired 
equalizer  in  any  mixer  position.  A  neon  volume  indicator  and  an 
electric  clock  are  mounted  in  the  line  of  screen  vision  on  the  front  of 
the  console.  Three  remote  controls  for  changing  the  location  of  the 


X2   NAILIN&    STRIPS 


FIG.  9.     Detail  of  monitoring  room;    stage  obser- 
vation window. 


"breakaway  point"  (post)  of  the  electronic  compressor  amplifiers  are 
installed  conveniently  near  the  dialog  mixer.  Remote  volume  con- 
trol of  the  projection  reproducing  system  is  provided  on  the  console. 
A  projected  footage  counter  is  placed  below  the  picture  screen,  and  a 
reset  button  is  located  on  the  dubbing  console. 

Scoring  Monitor  Room. — Special  consideration  was  given  to  the  de- 
sign of  the  scoring  monitoring  room,  which  is  usually  a  small  room 
ranging  in  volume  from  500  to  5000  cubic-feet.  Adjunct  to  a  more 


368  LOOTENS,  BLOOMBERG,  AND  RETTINGER    [j.  s.  M.  P.  E. 

voluminous  and  expensive  stage,  they  do  not  always  receive  the  same 
attention  during  the  design  period  that  the  larger  enclosure  receives, 
and  often  are  accorded  only  such  space  as  fits  in  conveniently  with  the 
grounds.  They  are  important  rooms,  however,  since  arrangement  of 
orchestra  and  tonal  balance  are  usually  regulated  by  the  mixer  in 
these  rooms. 

It  is  well  known  that  enclosures,  the  dimensions  of  which  are  not 
large  compared  with  the  wavelength  of  the  sound,  exhibit  a  phenome- 
non known  as  room  resonance.  Under  such  condition  one  or  more 
of  the  lower  modes  of  vibration  of  the  room  may  be  prominently 
stimulated,  causing  intensification  of  the  low-frequency  components 
of  the  sound  emanating  from  the  monitoring  speaker.  It  is  often 
thought  that  by  providing  irregularities  in  the  room,  such  as  pilasters 
or  corrugations,  or  by  making  the  enclosure  non-rectangular,  room 
resonance  can  be  eliminated.  Such,  of  course,  is  not  the  case.  It  is 
more  difficult  to  determine  theoretically  the  frequencies  of  the  eigen- 
tones  or  damped  free  vibrations  in  a  non-rectangular  room,  but  the 
number  of  them  occurring  within  a  certain  frequency-interval  is  cer- 
tainly not  increased.  For  that  reason,  also,  cubical  rooms,  or  even 
rooms  having  two  dimensions  alike,  are  less  to  be  recommended  than 
oblong  rooms,  since  their  eigentones  may  superimpose,  causing  an 
even  greater  intensificiation  of  a  certain  frequency  in  the  room.  It 
is  the  number  of  resonant  vibrations  within  a  given  frequency  band 
that  is  important  in  a  room,  since  the  larger  this  number  the  more 
will  every  forced  vibration  in  that  interval  coincide  with  a  natural 
mode  of  vibration  of  the  room.  If  the  room  is  not  too  small,  one  can 
calculate  the  number  of  eigentones  within  a  band  by  the  Rayleigh- 
Jeans  formula6  for  the  optical  case.  This  is  : 


where  F  =  frequency. 

V  =  volume  of  room. 

c  =  velocity  of  sound. 

N  =  number  of  eigentones  in  the  interval  ranging  from  F  to  F  +  A  F. 

This  equation  shows  that  the  number  of  eigentones  within  a  certain 
frequency-band  is  proportional  to  the  volume  of  the  room,  so  that 
pilasters,  corrugations,  etc.,  exert  no  effect  in  changing  this  number 
except  so  far  as  the  room  is  made  smaller  by  them  —  an  undesired  con- 
dition. 


April,  1939] 


DUBBING  AND  SCORING  STAGE 


369 


For  this  reason  the  scoring  monitoring  room,  of  4500  cubic-feet 
volume,  was  kept  rectangular  in  shape  with  dimensions  of  9  feet  high, 
18  feet  wide,  and  28  feet  long  (ratio  1:2:3)  (Figs.  1,  2).  The  moni- 
toring room  floor  is  covered  with  a  hair-felt  pad  and  a  heavyweight 
broadloom  carpet.  The  walls  are  treated  with  1/z-inch  fiber  insula- 
tion board  applied  directly  over  1  inch  of  cement  plaster.  The  ceiling 
is  of  plaster  board,  1/2-inch  brown  plaster  finished  with  1/2-inch 
"Acoustite"  plaster. 

Another  important  property  of  a  monitoring  room  is  adequate 
sound  insulation  between  it  and  the  adjoining  studio.  The  wall  ad- 
jacent to  the  stage  is  structurally  separated  from  the  stage  wall  and  is 


INPUT.  .    JJJPUT  INPUT  Ouw^  arfpur*  ixpur  INPUT  INPUT 

FIG.  10.     Diagram  of  scoring  mixer. 

supported  by  a  separate  footing.  The  matter  of  sufficient  insulation 
through  the  observation  window  deserves  particular  attention. 
Such  windows  often  consist  of  two  panes  of  glass  separated  by  a  small 
air  space  (1  to  3  inches)  with  one  sheet  of  glass  sometimes  thicker 
than  the  other.  Work  by  J.  E.  R.  Constable7  has  shown  that  such 
an  arrangement  is  invariably  less  insulative  for  the  lower  frequencies 
than  a  single  pane  having  the  combined  thickness  of  the  two  panes. 
Increased  insulation  by  means  of  two  sheets  of  glass  is  achieved  only 
if  the  panes  are  at  least  4  inches  apart,  and  when  the  wall  space  be- 
tween the  panes  is  treated  with  a  sound-absorbing  material  such  as 
fiberboard  or  acoustic  plaster. 

The  observation  window  (Fig.  9)  for  this  monitoring  room  consists 
of  two  double  panes  of  y4-inch  plate-glass,  with  one  pair  inclined  to 
the  other  at  a  small  angle  and  the  pairs  separated  from  each  other  as 


370 


LOOTENS,  BLOOMBERG,  AND  RETTINGER    [J.  S.  M.  p.  E. 


shown.  The  wall  space  between  the  pairs  of  sheets  is  treated  with 
V2-inch  nberboard,  and  the  rebate  around  the  edges  of  the  pane  is 
closed  tightly  against  the  pane  with  weather-stripping  between  rebate 
and  glass. 

The  equipment  in  the  scoring  monitoring  room  consists  of  a  6-posi- 
tion  mixer  console  and  a  RCA  two-way  theater  loud  speaker  system. 
The  mixer  circuit  (Fig.  10)  utilizes  a  switching  arrangement  for  di- 
viding the  six  mixers  into  two  groups  of  three,  connected  to  two  sepa- 
rate recording  channels.  A  soloist  and  an  orchestra  accompaniment 
if  isolated  with  acoustic  baffles  can  be  recorded  simultaneously  on  two 
separate  films.  Whenever  the  channel  is  used  as  described,  the  scor- 
ing monitor  is  connected  to  both  recording  systems.  A  neon  volume 


FIG.  11.     Amplifier  rack;     Bays  1  to  7. 


indicator  and  a  high-speed  Weston  meter  volume  indicator  are  used  as 
level  indicators.  High-  and  low-frequency  attenuation  is  supplied  in 
four  mixer  positions.  A  patch  bay  is  furnished  with  the  necessary 
trunks  to  the  dubbing  console  amplifier  bays  and  associated  equip- 
ment. 

A  signal  and  PA  panel  is  provided  with  the  additional  feature  of  a 
separate  PA  switch  connecting  directly  with  a  PA  microphone  and 
speaker  on  the  music  director's  stand  enabling  the  mixer  to  converse 
privately  with  the  music  director.  A  remote  variable  jEZ"-pad  control 
(Figs.  13,  15)  is  provided  to  raise  or  lower  the  "breakaway  point"  of 
No.  4  compressor  amplifier  (post)  used  in  the  scoring  channel.  A 
reset  button  for  the  projected  footage  counter  is  conveniently  located 
on  the  console. 


April,  1939] 


DUBBING  AND  SCORING  STAGE 


371 


Ventilation  and  Heating. — The  ventilating  and  heating  system  con- 
sists of  niters,  fan,  and  gas  heater  all  mounted  on  cork  insulation  pads 
on  the  roof  adjacent  to  the  projection  room.  In  order  to  reduce  wind 
noises  and  fan  vibration,  a  fan  normally  rated  at  20,000  cfm.  is 
driven  at  a  speed  to  produce  6000  cfm.  The  duct  leading  from 
the  heater  to  the  discharge  into  the  rooms,  is  lined  with  2-inch  rock- 
wool  blankets,  the  surface  of  which  is  covered  with  a  double  thickness 
of  40-44  muslin.  Staggered  acoustic  absorbing  panels  of  2-inch  rock- 
wool  are  placed  at  right  angles  to  the  air  travel  at  2-ft.  intervals 
throughout  the  length  of  the  duct. 
The  discharge  openings  from  the 
monitoring  room  and  stage  are 
designed  so  that  the  maximum 
velocity  of  the  air  does  not  exceed 
300  feet  per  minute.  To  prevent 
external  noise  from  entering  the 
room,  acoustic  insulating  panels  of 
1/2-hich  fiber  insulation  board  are  in- 
stalled at  right  angles  to  the  air-flow. 

Machine  Room. — The  machine 
room  (Fig.  1)  contains  the  amplifier 
racks,  recorders,  film  phonographs, 
sound-heads,  and  portable  acetate 
recorder.  Two  complete  recording 
channels  are  provided  either  for 
scoring  or  dubbing.  There  are 
eight  amplifier  bays  containing  the 
following  equipment : 

Bay  No.  1  (Fig.  11)  contains  six 
microphones  and  eight  photocell 

pre-amplifiers  mounted  in  the  upper  portion  of  the  rack.  The  micro- 
phone pre-amplifier  is  a  two-stage  amplifier  with  an  overall  gain  of 
47  db.  at  1000  cycles.  The  first  stage  of  this  amplifier  utilizes  the 
new  RCA  1603  tube,8  fed  by  a  specially  designed  input  transformer. 
Equalization  for  film-transfer  losses  is  incorporated  in  these  pre- 
amplifiers and  consists  of  a  rising  characteristic  starting  from  1000 
cycles  and  rising  to  6  db.  at  6000  cycles.  The  phototube  pre-ampli- 
fier is  also  a  two-stage  amplifier  using  the  RCA  1603  in  the  first  stage. 
The  overall  gain  is  43  db.  at  1000  cycles.  Equalization  for  re-record- 
ing transfer-losses  is  incorporated  in  these  pre-amplifiers.  Both  types 


FIG.     12.      Mountings    for    micro- 
phone and  phototube  pre-amplifiers. 


372 


LOOTENS,  BLOOMBERG,  AND  RETTINGER    [j.  s.  M.  p.  E. 


of  pre-amplifiers  are  mounted  in  rectangular  cases  which  fit  in  shelves 
at  the  rear  of  the  rack  (Fig.  12).  The  male  output  receptacle  at  one 
end  of  the  amplifier  case  connects  with  the  female  receptacle  on  the 
rack,  and  the  input  receptacle  on  the  amplifier  is  connected  by  a 
short  piece  of  cable  from  the  rack.  Power  switches,  metering  jacks, 
and  meter  are  supplied  on  the  front  face  of  the  rack. 

A  section  of  jack  rows,  immediately  below,  contains  high-  and  low- 
level  trunks  to  the  dubbing  and  scoring  consoles,  the  circuit-test  lab., 
the  inputs  and  outputs  of  two  electronic  compressor  amplifiers,  and 
the  fixed  pads  and  variable  controls  associated  with  the  operation  of 


o 

00* 

A  WC 

<ETE 

RER 
' 

OE 

BLI 
XJB8 

LEC 

:  w 

NG 

ION 

3DUC 
>TAG 

c 

TION 
E 

V 

S   IN 
/ 

C. 

S* 

-2 

/ 

B 

-^ 

*£~ 

*^~~ 

4 

^ 

* 

s* 

-6 

/ 

^ 

S" 

-8 

/ 

<s 

yS* 

•10 

-OUTPUT    LEVC 
IN   D» 

7 

^ 

/ 

8. 
C. 

D. 
E. 

IK    I 

COMP 

COMP 

UNCO 
BRCA 
BREA 

Bi 

RESS 

2 

"cs: 

UPRl 
KAW* 
KAVW 

OR    S 
0    DE 
OR     f 
1     &[ 

SSM 

V     ft 
Y     PC 

10.2 

INT 

D. 
c. 

INT 

INT 

CLO 
0    It 

4 

7    D 

or 
or 

ING 

ol 

0»f 
OMP 

TIU 

CL 
o  3 

css 

ESS 

0. 

>SIN 
DB. 

••R   M 

2R    N 

>   SEC. 

•14 

/\ 

.    TIME 

-16 

/ 

.11             / 

k.1  A  3 

/ 

N=U 

LE 

VEL 

FIG.  13.     Compressor  curves. 


the  compressors.     The  compressors  and  pads  are  mounted  below  the 
jack  rows. 

Compressor  amplifier  No.  1  is  designed  to  work  in  the  music-dub- 
bing channel.  The  compressor  consists  of  a  two-stage  push-pull  am- 
plifier, the  first  stage  of  which  is  a  pair  of  variable-mu  tubes  6K7,  the 
gain  of  which  is  controlled  by  the  output  of  the  rectifier  (6H6). 
The  rectifier  is  fed  by  a  one-stage  amplifier  (6C5)  which  is  bridged 
across  the  output  of  the  compressor.  The  harmonic  distortion  intro- 
duced by  the  compressor  at  an  output  of  plus  10  db.  is  less  than  1  per 
cent.  The  frequency-response  is  flat  within  plus  or  minus  l/z  db. 
from  30  cycles  to  10,000  cycles.  The  timing  characteristics  used  in 
this  compressor  are  two  milliseconds  for  operating  and  approximately 
100  milliseconds  for  release.  The  operating  characteristic  is  such 


April,  1939]  DUBBING  AND  SCORING  STAGE  373 

that  it  requires  a  17-db.  increase  of  input  to  raise  the  galvanometer 
the  final  3  db.  of  modulation  (curve  B,  Fig.  13).  This  compressor  is 
operated  in  this  manner  to  prevent  peak  voltages  from  overloading 
the  sound-track  without  affecting  the  normal  volume  changes  in  the 
music. 

No.  2  compressor  amplifier  is  designed  to  work  in  the  dialog  dub- 
bing channel  (Fig.  14).  The  timing  characteristics  used  are  2  milli- 
seconds for  operating  and  approximately  500  milliseconds  for  release. 
The  operating  characteristic  is  such  that  it  requires  a  20-db.  increase 
of  input  to  raise  the  galvanometer  the  final  10  db.  of  modulation,  as 
illustrated  by  curve  A  of  Fig.  13.  A  variable  T-pad  in  1-db.  steps  in 
the  output  of  the  compressor  is  provided  on  the  dubbing  console  and 
is  used  to  raise  or  lower  the  "breakaway  point." 

Bays  Nos.  2  and  3  (Fig.  11)  each  contains  a  complete  recording 
amplifier  channel  with  a  balanced  low-pass  filter,  45-cycle  high-pass, 
recording  amplifier,  and  d-c.  bridging  amplifier  (Figs.  14  and  15). 
The  balanced  low-pass  filter  has  variable  cut-off  frequencies.  Speech 
recording  is  done  with  the  6500-low-8000,  and  music  recording  with 
the  8000-low-10,000.  The  45-cycle  high-pass  with  operating  charac- 
teristics of  40-high-60  is  inserted  in  the  recording  channel  for  both 
music  and  speech  recording. 

The  108  recording  amplifier  is  used  as  the  main  gain-amplifier  and 
serves  to  bring  the  signals  from  the  low-level  mixer  circuits  to  a  zero 
level  at  the  bridging  bus.  It  employs  four  stages  of  amplification 
and  has  an  overall  gain  of  84  db.  at  1000  cycles.  The  frequency- 
response  is  flat  from  30  cycles  to  10,000  cycles,  within  ±  l/z  db.  A 
third  compressor,  which  has  the  operating  characteristics  of  compres- 
sor No.  1,  is  inserted  between  the  108  amplifier  output  and  the  bridg- 
ing bus,  preventing  excessive  peak  voltages  from  overloading  the 
galvanometer.  A  three-stage  bridging  amplifier  completes  the  cir- 
cuit between  the  bridging-bus  and  the  galvanometer.  The  overall 
gain  of  this  amplifier  is  40  db.  The  frequency  response  is  flat  from  30 
cycles  to  10,000  cycles  within  ±  y2  db.  All  these  amplifiers  are  pro- 
vided with  metering  jacks  and  a  meter  for  measuring  operating  volt- 
ages. 

Jacks  are  provided  in  both  bays  for  the  input  and  output  of  each 
individual  amplifier  and  its  associated  equipment  and  also  high  and 
low-level  trunks  to  the  different  bays  and  the  dubbing  and  scoring 
consoles.  The  noise  level  of  the  overall  channel  at  the  recording  gal- 
vanometer is  maintained  to  —60  db.  below  "O"  db.,  and  the  overall 


374 


LOOTENS,  BLOOMBERG,  AND  RETTINGER    [j.  s.  M.  p.  E. 


distortion  at  400  cycles  is  0.4  per  cent  at  100  per  cent  galvanometer 
deflection. 

Bay  No.  4  (Fig.  11)  is  the  circuit-test  lab.  and  contains  an  oscilla- 
tor, distortion-factor  meter,  gain  set,  and  a  patch-bay  containing 


n- 


FIG.  14.     ( Upper)  Diagram  of  dubbing  channel. 
FIG.  15.     (Lower)  Diagram  of  scoring  channel. 

trunks  to  all  bays.  The  jack  bay  provides  connections  to  various 
loss  pads,  transformers,  band-pass  niters  and  trunk  terminations  use- 
ful in  making  transmission  runs.  A  test-panel  at  the  lower  section  of 
the  bay  is  directly  connected  to  a  test-panel  in  the  truck-testing 


April,  1939] 


DUBBING  AND  SCORING  STAGE 


375 


platform  with  high-  and  low-level  trunks.  It  also  contains  power- 
supply  and  provision  for  checking  pre-amplifiers.  Connections  for  a 
portable  PA  system  or  a  telephone  hand-set  are  provided  for  com- 
munication between  the  truck  platform,  maintenance  room,  and  cir- 
cuit lab. 

Bay  Nos.  5  and  6  (Fig.  11)  contain  the  monitoring  amplifiers  and 
associated  equipment  used  with  the  scoring  and  dubbing  channels. 
Each  monitor  system  consists  of  a  40-watt  amplifier  fed  by  a  three- 
stage  bridging  amplifier  the  input  of  which  is  connected  across  the 
bridging  bus.  The  monitoring  equalization  is  connected  between 
these  two  amplifiers.  The  bridging  amplifier  is  similar  in  operating 


FIG.    16.     Noise  reduction  amplifier — Bay  8,  recorders, 
and  motor  switching  panel. 


characteristics  to  those  described  in  Bays  2  and  3.  The  output  of 
the  power  amplifier  feeds  the  two-way  speaker  system  through  its 
dividing  network.  The  total  harmonic  distortion  in  the  monitoring 
system  is  less  than  1  per  cent  at  a  power  output  of  40  watts. 

Bay  No.  7  (Fig.  11)  contains  the  PA  amplifiers  and  signal  systems. 
The  PA  system  includes  5  stations:  viz.,  recorder  No.  1,  recorder  No. 
2,  scoring  console,  dubbing  console,  and  projection  room. 

Two  voltage  and  two  power-amplifiers  supply  amplification  for  the 
PA  system.  A  PA  switching  panel  is  provided  for  making  the  con- 
nections to  and  from  the  selected  stations.  A  signal  patch  panel 
makes  it  possible  to  select  any  station  for  controlling  the  signal  and 
warning-light  system.  A  portable  PA  amplifier  is  provided  for  use 
with  a  portable  PA  microphone  and  speaker  which  can  be  used  for 


376  LOOTENS,  BLOOMBERG,  AND  RETTINGER    [J.  S.  M.  p.  E. 

test  communication.  A  patch-bay  is  furnished  containing  the  inputs 
and  outputs  of  these  amplifiers  with  trunks  to  the  other  bays. 

Bay  No.  8  contains  the  noise-reduction  amplifiers  (Fig.  16)  asso- 
ciated with  the  recording  channels,  mounted  between  the  two  record- 
ing machines.  These  noise-reduction  amplifiers  contain  new  de- 
velopments in  operating  characteristics.  The  timing  characteristics 
used  are  19  milliseconds  opening  and  approximately  220  milliseconds 
closing.  The  frequency  characteristic  of  the  amplifier  is  flat  from  30 
to  10,000  cycles  within  ±  l/2  db. 

A  meter  and  metering  jacks  are  provided  for  measuring  operating 
voltages.  A  patch-bay  is  furnished  with  the  inputs  and  outputs  of 
the  amplifiers  and  trunks  to  the  circuit  lab.  and  recording-channel 
bays.  This  rack  contains  also  an  RCA  modulated  oscillator  for  use 
in  determining  optimum  negative  and  printing  densities  for  processing 
control.9 

The  machine  room  also  contains  three  film-phonographs,  five 
sound-heads,  two  film-recorders,  and  mobile  acetate  recorder.  The 
three  film-phonographs  are  of  the  RCA  magnetic-drive  type  and  are 
used  for  reproducing  master  dialog  and  music  tracks.  The  five 
sound-heads  are  of  the  RCA  rotary-stabilizer  type,  and  are  used  for 
sound-effects  tracks.  A  loop  arrangement  is  mounted  above  the 
sound-heads  providing  for  film  loops  up  to  300  feet  in  length.  All 
reproducers  are  equipped  with  switches  for  reproducing  either  stand- 
ard or  push-pull  tracks.  Automatic  motor-driven  rewinds  are  fur- 
nished on  all  reproducers  so  that  it  is  unnecessary  to  remove  the  film 
from  the  machines  for  rewinding.  An  inspection  bench  is  provided 
with  motor-driven  rewinds  and  illuminated  inspection  plates. 

In  the  rear  of  the  machine  room  is  a  panel  with  outlet  for  a  mobile 
"acetate"  recorder  which  is  self-contained  in  a  steel  cabinet  mounted 
on  pneumatic  tires.  The  mechanism  is  provided  with  both  selsyn 
and  synchronous  motors  driving  the  turntable  at  a  speed  of  78  rpm. 
An  RCA  72A  cutting  mechanism,  the  new  RCA  pick-up,  tone-arm, 
associated  filters  and  equalizer,  microscope,  and  record-spotting  mecha- 
nism complete  the  operating  accessories.  The  amplifying  equip- 
ment consists  of  an  amplifier  system  and  associated  filament  rectifier. 
A  meter  volume-indicator  is  provided  in  the  cutter  circuit.  The 
external  speaker  is  the  new  RCA  all-metal  exponential  horn  with  a 
power  handling  capacity  of  24  watts.  A  microphone  input  is  fur- 
nished so  that  the  mobile  unit  may  be  used  as  a  PA  system  on  the  pro- 
duction set.  For  immediate  playback  purposes  when  operating  with 


April,  1939]  DUBBING  AND  SCORING  STAGE  377 

the  recording  channel,  a  switch  on  the  acetate  control  panel  operates 
a  relay  connecting  the  acetate  reproducer  and  network  to  the  moni- 
toring two-way  horn  system  in  both  the  scoring  stage  and  scoring 
monitoring  room. 

Two  RCA  ultraviolet  film  recorders  of  the  magnetic-drive  type  are 
equipped  with  a  photographic  slating  device  and  film-punch  operated 
simultaneously  by  a  hand-lever.  An  exposure  meter,  corrected 
for  temperature  variations  and  which  gives  accurate  photometric 
readings  for  controlling  exposure,  is  mounted  on  the  front  of  the  opti- 
cal systems.  The  recorders  are  mounted  on  tables  (Fig.  16)  the 
front  of  which  contains  a  control  panel  with  a  PA  microphone,  tele- 
phone hand-set,  interlock  control  switches,  signal-control  switches, 
and  signal  lights. 

Alongside  the  recording  tables  on  the  front  wall  (Fig.  16)  is  the 
selsyn  distributor  control  panel  which  contains  twelve  specially  de- 
signed six-pole  double-throw  switches,  enabling  the  operator  to 
switch  any  or  all  of  the  reproducer  and  projection  interlock  motors 
from  one  distributor  to  the  other.  A  convenience  outlet  is  installed 
near  each  reproducer  to  provide  a  connection  for  variable-speed  shots. 
A  remote  portable  variable-speed  control  is  provided  for  operation 
near  the  dubbing  console. 

Projection  Booth. — The  projection  room,  which  is  located  above  a 
section  of  the  hallway  and  one  end  of  the  monitor  room  (Figs.  1,  2) 
is  supported  by  columns  which  are  structurally  separated  from  the 
stage  and  the  monitor  room  walls  and  ceiling.  The  projection  room 
floor  consists  of  a  2-inch  layer  of  asphalt  on  top  of  which  is 
cemented  IVVinch  cork  insulation  pads.  One-quarter  inch  cork 
carpet  is  cemented  to  the  top  of  the  cork  pads.  This  construction 
eliminates  the  possibility  of  footfalls  being  heard  in  the  monitoring 
room  and  stage.  The  ceiling  and  the  upper  3  feet  of  the  projection 
room  walls  are  finished  with  y2-in.  "Acoustite"  plaster. 

The  projection  booth  contains  two  Simplex  E-7  projection  heads 
with  removable  aperture  plates,  two  high-intensity  Peerless  Magnarcs 
and  RCA  PG118  double  sound-head  projection  system  with  one  head 
driven  by  a  synchronous  motor  and  the  second  head  by  a  selsyn  mo- 
tor, and  two  RCA  preview  attachments.  Both  heads  are  push-pull 
or  standard.  Each  arc  lamp  is  supplied  by  a  General  Electric  cop- 
per-oxide rectifier  of  65-ampere  capacity.  The  amplifier  rack  con- 
tains a  patch-bay  with  the  inputs  and  outputs  of  the  voltage  and 
power  amplifiers,  and  trunk  lines  to  the  dubbing  and  scoring  consoles 


378  LOOTENS,  BLOOMBERG,  AND  RETTINGER    [J.  S.  M.  P.  E. 

and  the  amplifier  bays.  A  volume  indicator  meter  is  mounted  on  the 
panel  and  is  connected  by  a  switch  and  transformer  across  the  output 
to  the  stage  speakers  to  check  the  level  and  make  frequency  runs. 
The  photocell  output  of  either  sound-head  can  be  plugged  into  photo- 
cell pre-amplifier  No.  9 — Bay  1,  in  the  machine  room,  through  a  5-pin 
Cannon  receptacle  mounted  on  the  wall  in  front  of  the  sound-heads. 
A  signal  telephone  and  PA  panel  are  mounted  on  the  wall  between 
the  two  projectors.  A  PA  and  monitor  speaker  are  mounted  in  the 
end  of  the  booth.  A  motor-driven  rewind  cabinet,  a  film-storage 
cabinet  and  an  inspection  table  with  hand  rewinds  completes  the 
equipment.  On  the  rear  wall  is  a  remote  switch  which  turns  on  all 
the  rectifiers  in  the  power  room  supplying  the  projection  amplifiers, 
exciter  lamps,  and  field  supplies.  A  table-switch  on  each  projector 
controls  the  relay  supplying  alternating  current  to  the  arc  rectifiers. 

Power  Room. — The  power  room  contains  all  the  rectifiers  supplying 
the  filament  and  plate  voltages  of  all  recording  and  reproducing  am- 
plifiers, exciter  lamp  supplies,  and  field  supplies.  Two  selsyn  dis- 
tributors and  an  emergency  truck-battery  charging  generator  are  also 
located  in  the  power  room.  Control  panels,  starting  relays,  and  all 
associated  switches  are  mounted  conveniently  on  the  wall.  The 
rectifiers  are  mounted  in  specially  built  racks  on  one  side  of  the  power 
room,  and  all  the  wiring  is  contained  in  easily  accessible  gutters. 
Telephone  or  a  portable  PA  outlet  provides  communication  to  the 
truck  platform,  maintenance  room,  projection  booth,  and  machine 
room. 

Maintenance  Room. — The  maintenance  room  (Fig.  1)  contains  a 
work-bench  along  two  side-walls,  with  cupboards,  drawers,  and  cabi- 
nets for  storing  accessories  and  spare  parts.  A  test  and  patch-panel 
is  located  above  the  work-bench  with  filament  and  plate  supply  for 
any  amplifier  tests  and  high  and  low-level  trunks  to  the  truck  plat- 
form and  the  circuit  lab.  telephone;  and  portable  PA  communication 
with  the  truck  platform,  machine  room,  projection  booth  and  power 
room,  is  available.  Selsyn  interlock  and  220- volt  3 -phase  connections 
are  furnished  for  testing  motors.  A  complete  complement  of  testing 
instruments,  meters,  and  accessories  is  provided  for  all  testing. 

Recording  Truck  Platform. — The  truck  platform  contains  stalls  for 
four  sound  trucks.  In  each  stall  is  a  220-volt  3-phase  outlet  for  sup- 
plying the  tungar  battery-chargers  in  each  truck.  Each  stall  con- 
tains a  16- volt  and  an  8- volt  d-c.  charging  outlet,  supplied  by  a 
generator  in  the  power  room  for  giving  the  16-  and  8- volt  batteries  in 


April,  1939]  DUBBING  AND  SCORING  STAGE  379 

the  sound- trucks  an  emergency  boost  charge.  In  the  center  rear  wall 
of  the  truck  platform  is  a  test-panel  with  facilities  for  making  audio, 
continuity,  and  megger  tests  of  all  cables  used.  A  patch-bay  in  the 
test-panel  provides  high-  and  low-level  lines  to  the  maintenance  room 
or  the  circuit  test  lab.  Thus  any  truck  amplifier  can  be  patched  to 
the  circuit- test  lab.  for  frequency  runs.  Telephone  or  portable  PA 
communication  is  provided  between  the  sound-trucks  or  the  truck 
test-panel,  maintenance  room,  power  room,  and  circuit- test  lab. 

REFERENCES 

1  RETTINGER,  M.:    "Note  on  Reverberation  Characteristics,"  /.  Acoust.  Soc. 
Amer.,  VI  (July,  1934),  No.  1,  p.  51. 

2  KNUDSON,  V.  O.:    "Recent  Developments  in  Architectural  Acoustics,"  Rev. 
Modern  Physics,  VI  (Jan.,  1934),  No.  1,  p.  14. 

3  RYRING,  C.  F.:   "The  Reverberation  Time  in  Dead  Rooms,"  /.  Acoust.  Soc. 
Amer.,  1  (1930),  No.  2,  p.  217. 

4  MACNAIR,  W.  A.:     "Optimum  Reverberation  Time  for  Auditorium,"   J. 
Acoust.  Soc.  Amer.,  1  (1930),  No.  3,  p.  242. 

6  STRUTT,  M.  J.  O.:    "Raumakustick."     Handbuck  der  Experimental  physik 
von  Wien-Harms,  Bd.  17,  Techn.  Akustick  (1934),  S.  460. 

•JEANS,  J.  H.:  "On  the  Partition  of  Energy  between  Matter  and  Ether," 
Phil.  Mag.,  10  (1905),  No.  91. 

7  CONSTABLE,  J.  E.  R. :  Phil.  Mag.,  18  (1934),  No.  7,  p.  321. 

8  HOLLANDS,  L.  C.,  AND  GLOVER,  A.  M. :   "Vacuum  Tube  Engineering  for  Mo- 
tion Pictures,"  J.  Soc.  Mot.  Pict.  Eng.,  XXX  (Jan.,  1938),  No.  1,  p.  42. 

9  BAKER,  J.  O.,  AND  ROBINSON,  D.  H.:   "Modulated  High-Frequency  Record- 
ings as  a  Means  of  Determining  Conditions  for  Optional  Processing,"  /.  Soc.  Mot. 
Pict.  Eng.,  XXX  (Jan.,  1938),  No.  1,  p.  3. 

DISCUSSION 

MR.  MACNAIR:  The  expression  "closing  time"  is  used  with  respect  to  the 
compressor.  Exactly  what  does  that  mean? 

MR.  WOLFE  :  A  more  nearly  correct  term  would  be  "restoring  time."  What  was 
meant  was  obviously  the  time  required  for  the  gain  of  the  compressor  to  restore 
to  its  normal  conditions.  The  term  "closing  time"  was  borrowed  from  noise- 
reduction  terminology. 

MR.  MACNAIR:  Sooner  or  later,  in  working  with  noise-reduction  units  and 
compressors,  we  must  agree  on  these  terms.  Theoretically,  the  circuit  never 
restores  except  in  an  infinity  of  time,  and  some  of  us  choose  to  talk  about  1/e, 
because  we  used  to  be  scientists.  Others  talk  about  the  99-per  cent  point,  and  so 
on.  I  think  sooner  or  later  we  must  agree  on  something. 

MR.  WOLFE  :  In  our  company  we  are  trying  to  standardize  on  the  90-per  cent 
point,  as  the  point  at  which  the  restoring  or  closing  time  is  measured. 

MR.  KELLOGG:  I  have  often  wondered  what  difference  in  impression  you  get 
in  a  small  room  as  compared  with  a  large  room  if  both  of  them  have  the  same 


380  LOOTENS,  BLOOMBERG,  AND  RETTINGER 

reverberation  time  for  all  frequencies.  That  is  theoretically  quite  possible,  and 
no  doubt  it  is  quite  often  moderately  well  approached.  I  do  not  believe  many  of 
us  would  be  fooled  into  thinking  we  had  a  larger  room,  but  I  have  never  seen  the 
point  brought  out  quite  as  well  as  I  think  the  authors  do. 

The  authors  mention  the  fact  that  the  small  room  has  a  shorter  mean  free  path 
for  the  sound,  and  therefore  the  echoes  come  much  more  frequently.  You  get 
many  more  echoes  in  the  same  time,  and  if  the  absorption  is  adjusted  so  the  total 
reverberation  time  is  the  same,  then  the  dying  away  of  the  sound  takes  place  in 
smaller  jumps.  In  other  words,  the  sound  is  chopped  up  a  great  deal  finer. 

It  is  of  some  interest  that  with  a  large  orchestra  we  seem  to  want  the  rather 
slower  chopping  that  we  can  get  with  the  large  room.  With  a  smaller 
orchestra  it  is  obvious  that  since  there  are  fewer  instruments  you  might  need 
more  chopping  up  of  the  echoes.  In  other  words,  the  large  number  of  players 
does  very  much  in  one  way  what  the  frequent  echoes  do  in  the  other.  That  may 
have  something  to  do  with  our  preference  for  a  large  room  where  we  have  a  large 
number  of  players. 

MR.  BUTLER:  I  believe  distribution  of  frequency  in  a  room  is  the  answer  to 
Mr.  Kellogg's  question.  If  we  have  a  definite  reverberation  from  one  end  of  the 
room  to  the  other  and  can  distribute  it,  it  is  all  right;  but  if  we  can  distribute  the 
frequent  small  echoes,  I  feel  we  would  have  a  better  overall  condition. 


UNIDIRECTIONAL  MICROPHONE  TECHNIC* 

J.  P.  LIVADARY**  AND  M.  RETTINGERt 

Summary. — After  describing  the  constructon  of  a  unidirectonal  microphone  several 
desirable  factors  are  discussed  connected  with  the  use  of  such  a  transmitter  for  recording 
sound  in  motion  picture  studios.  Six  illustrations  show  how  the  microphone  may  be 
used  to  advangate  under  specific  set  conditions,  and  four  diagrams  illustrate  its  use 
in  recording  various  types  of  music. 

A  unidirectional  microphone  is  a  microphone  that  is  operated 
partly  by  sound  pressure  and  partly  by  a  pressure  gradient,  or  differ- 
ence, in  sound  pressure  between  the  two  sides  of  the  part  so  driven. 
The  moving  element  in  the  most  commonly  used  unidirectional  micro- 
phone is  a  light  corrugated  metallic  ribbon  suspended  in  the  magnetic 
field  of  a  permanent  magnet  and  divided  into  two  parts.  One  section, 
freely  accessible  to  air  vibrations  from  both  sides,  acts  as  a  so-called 
velocity  microphone,  because  the  induced  emf .  in  the  ribbon  is,  within 
practical  limits,  directly  proportional  to  the  particle  velocity  of  the 
driving  sound-wave.  The  other  section  of  the  ribbon,  exposed  to 
pressures  from  sound-waves  on  one  side  only,  has  its  other  side  termi- 
nated into  an  acoustic  labyrinth,  and  responds  as  a  pressure-operated 
device,  the  induced  emf.  in  the  ribbon  being  proportional  to  the 
pressure  of  the  driving  sound-wave.  Because  the  ribbons  are  in 
series,  the  combined  output  from  the  microphone  is 

Eu  =  Ep  +  Evcos8  (1) 

where  Eu  =  Voltage  output  of  the  unidirectional  microphone  for  sound  origi- 
nating in  the  direction  6. 

Ep  =  Voltage  output  of  pressure-driven  element  for  sound  originating 
in  any  direction. 

Ev  =  Voltage  output  of  pressure-gradient  element  for  sound  originating 
in  the  direction  6  =  0,  which  corresponds  to  the  direction  of  a 
normal  to  the  flat  part  of  the  ribbon. 

*  Presented  at  the  1938  Fall  Meeting  at  Detroit,  Mich. ;  received  September 
23,  1938. 

**  Columbia  Pictures  Corp.,  Ltd.,  Hollywood,  Calif. 
t  RCA  Manufacturing  Co.,  Inc.,  Los  Angeles,  Calif. 

381 


382  J.  P.  LlVADARY  AND  M.  RETTINGER  [J.  S.  M.  P.  E. 

Assuming  that  emf.  generated  by  each  ribbon  is  adjusted  to  be  the 


same  for  zero  degree  incidence  (6  =  0),  then  EP  =  Ev 

=  Eo  (1  +  cos  8) 


E0)  and 


(2) 


The  directional  characteristic  of  the  unidirectional  microphone  as 
expressed  by  eq.  2  is  a  cardiod  of  revolution,  with  the  axis  of  revolu- 
tion normal  to  the  plane  of  the  ribbon.  This  characteristic  is  shown 
in  Fig.  1  for  the  two-dimensional  case. 


FIG.     1.     Directional     characteristic     of 
unidirectional  microphone. 

Four  definite  advantages  for  sound  recording  are  expressed  by  such 
a  directional  characteristic. 

The  fact  that  the  sound  striking  the  microphone  from  the  rear  is 
greatly  attenuated  prevents  recording  such  undesirable  sounds  as 
camera  noise,  back-stage  reflections,  and  what  incidental  noises  might 
occur  from  that  direction  during  a  recording. 

The  large  solid  angle  of  reception  over  which  the  microphone  re- 
ceives sound  without  appreciable  attenuation  indicates  that  practically 
any  action  can  be  covered  with  a  single  microphone.  This,  of  course, 
eliminates  what  interference  is  produced  by  the  microphones  when 
two  or  more  units  are  used  in  covering  an  action,  since  the  identical 
results  are  produced  when  the  voltages  from  the  microphones  are  out 


April,  1939]  UNDIRECTIONAL  MICROPHONE  TECHNIC  383 

of  phase  as  when  sound-waves  unite  to  produce  near  or  complete 
cancellation  of  the  vibrations  in  space.  Much  smoother  dialog  may 
therefore  be  expected  with  the  use  of  one  unidirectional  microphone 
in  place  of  two  or  more  microphones  of  different  types. 

The  energy  response  of  the  unidirectional  microphone  to  sound 
originating  in  random  directions  is  one-third  that  of  a  nondirectional 
microphone.  This  means  that  for  the  same  allowable  reverberation, 
the  unidirectional  microphone  can  be  used  at  1.7  times  the  distance 
of  a  nondirectional  mirophone.1  No  loss  in  intelligibility  occurs  when 
this  greater  distance  is  used,  since  the  per  cent  syllable  articulation 
of  recorded  sound  is  dependent  only  upon  the  amount  of  recorded 
reverberation*  in  a  room  free  from  noise  and  echoes.2 

The  directional  characteristic  of  the  unidirectional  microphone  is 
independent  of  frequency  within  all  practicable  reception  angles.  One 
of  the  most  objectionable  qualities  of  a  pressure-operated  microphone 
used  in  recording  sound  is  the  directional  response  that  such  a  trans- 
mitter may  exhibit  for  frequencies  above  2000  cycles.  While  the 
polar  response-frequency  characteristic  of  an  ideal  pressure-operated 
microphone  is  a  circle,  as  far  as  a  plane  through  the  transducer  is  con- 
cerned, in  practice  the  response  at  the  higher  frequencies  becomes 
noticeably  attenuated  for  angles  larger  than  30  degrees  on  either  side 
of  the  normal,  or  zero  degree,  incidence  axis  through  the  microphone. 
While  such  undesirable  effects  can  be  reduced  by  recording  all  dialog 
at  some  angle  off  the  normal,  such  practice  is  greatly  dependent  on 
the  experience  and  skill  of  the  man  who  is  entrusted  with  the  handling 
of  the  transmitter  during  a  recording.  Considering,  moreover,  the 
manifold  situations  of  a  recording,  the  several  sources  of  sound  that 
either  simultaneously  or  in  quick  succession  must  be  recorded  with  a 
minimum  of  delay  and  hazard,  it  stands  to  reason  that  a  microphone 
having  a  directional  response  independent  of  frequency  will  greatly 
reduce  these  objectionable  abrupt  loudness  and  quality  variations. 

It  may  be  stated  at  this  point  that  a  pressure-operated  micro- 
phone, such  as  used  in  practice,  tends  toward  a  ratio  of  direct  to  re- 
flected sound  in  the  recording  that  is  larger  for  the  high  frequencies 
than  for  the  low  ones,  if  the  microphone  is  used  "beam  on,"  because 
reflected  high  frequencies  striking  the  microphone  at  large  angles  of 
incidence  are  attenuated  by  the  microphone.  While  this  has  a  ten- 


*  Recorded  reverberation  is  defined  as  the  ratio  of  generally  reflected  to  direct 
sound  energy. 


384  J.  P.  LWADARY  AND  M.  RETTINGER  [J.  S.  M.  P.  E. 

dency  to  lend  greater  "presence"  to  the  sound  recorded  b'y  a  pressure- 
operated  microphone,  the  increased  amount  of  low  frequencies  re- 
corded may  exert  a  masking  effect  on  this  sound.  Only  experience 
will  be  able  to  state  how  important  these  two  factors  are,  whether 
they  neutralize  each  other,  or  whether  the  absence  of  one  of  them  in 
the  unidirectional  microphone  will  be  noticed  in  recordings  made  with 
such  a  microphone. 

In  the  following  it  is  intended  to  describe  a  number  of  scenes  that 
lend  themselves  particularly  well  to  the  use  of  a  unidirectional  micro- 
phone. These  scones  are  specific,  but  by  no  means  infrequent,  and 
certainly  have  variations  that  can  be  covered  equally  well. 

Screening  of  Undesirable  Sounds. — Fig.  2  shows  a  subway  tunnel 
the  "ceiling"  of  which  consists  of  iron  grids  opening  to  the  sidewalk 
of  the  street  above.  It  is  intended  to  record  only  the  dialog  of  the  two 
persons  in  the  tunnel,  but  not  the  footfalls  of  the  passers-by  above. 
The  camera,  on  the  other  hand,  is  to  photograph  the  entire  scene,  the 
actors  speaking  as  well  the  actors  overhead.  A  unidirectional  micro- 
phone, skillfully  concealed  in  the  grating  and  oriented  with  the  "dead 
plane"  toward  the  sidewalk,  represents  an  ideal  solution  for  this 
difficult  scene,  since  footfalls  may  later  be  "dubbed  in"  to  an  extent 
that  will  not  mar  the  dialog  but  yet  permit  the  auditor  in  the  theater 
to  hear  actual,  if  faint,  footsteps. 

Fig.  3  shows  the  hull  of  a  ship  in  which  two  actors,  at  some  dis- 
tance from  one  end,  are  talking.  The  hull,  which  is  that  of  a  real 
boat,  is  very  "live,"  and  considerable  reflected  sound  comes  from  the 
end  being  photographed.  •  A  unidirectional  microphone  so  oriented 
that  its  zero-degree  incidence  axis  points  toward  the  camera,  will 
eliminate  much  of  the  undesirable  reflected  sound  from  the  pictured 
end. 

Fig.  4  shows  an  actor  being  photographed  near  a  tree  and  a  water- 
fall, where  the  rush  of  the  water  is  to  be  "dubbed  in"  in  re-recording. 
A  unidirectional  microphone  so  placed  that  the  180-degree  axis  is 
directed  toward  the  waterfall  will  reduce  the  sound  of  the  water  by 
approximately  20  db.,  which  represents  the  difference  in  response  be- 
tween the  front  and  the  rear  of  the  unidirectional  microphone. 

Use  of  Greater  Microphone  Distance. — Fig.  5  represents  a  medium 
shot.  Good  recording  practice  calls  for  an  "acoustic  perspective"  con- 
forming to  the  visual  perspective  of  the  picture  seen  on  the  screen.3 
This  means  that,  if  a  microphone  may  be  said  to  have  "focal  length," 
its  value  should  be  equivalent  to  the  optical  focal  length  of  the  camera 


April,  1939]          UNIDIRECTIONAL  MICROPHONE  TECHNIC 


385 


*K 

^ 


1! 

I 


fctic,ao«q«  panes 


386  J.  P.  LIVADARY  AND  M.  RETTINGER        [j.  s.  M.  P.  E. 

lens  in  order  to  achieve  effective  visual  and  aural  coordination.  In 
practice,  however,  it  is  not  only  difficult  in  many  cases  to  position  the 
microphone  so  as  not  to  be  within  the  camera  angle,  but  cameras  are 
often  equipped  with  lenses  of  greater  focal  length,  which  calls  for  an 
even  shorter  distance  between  the  speaker  and  the  microphone  than 
between  the  speaker  and  the  camera.  If  a  pressure  microphone  is 
placed  barely  outside  the  camera  angle  when  such  a  condition  exists, 
the  auditor  in  the  theater  may  gain  the  impression  that  the  source 
of  sound  is  not  on  the  screen  but  behind  it.  A  unidirectional  micro- 
phone, however,  having  similarly  a  large  "acoustic  focal  length," 
when  placed  at  the  position  of  the  pressure  microphone  will  create  the 
effect  of  a  pressure  microphone  situated  at  a  distance  six-tenths  of 
that  between  the  speaker  and  the  unidirectional  microphone.  Indeed, 
with  some  care  a  position  can  be  chosen  for  the  unidirectional  micro- 
phone that  will  not  only  maintain  a  high  degree  of  intelligibility  in  the 
recorded  sound  when  the  transmitter  is  placed  outside  the  view  of  the 
camera,  but  also  the  "acoustic  perspective"  can  be  made  fully  com- 
mensurate with  the  depth  of  the  image  on  the  screen — a  condition 
often  referred  to  as  "screen  presence." 

Directional  Response  Independent  of  Frequency. — Fig.  6  shows  two 
actors  engaging  in  rapid  diaglog.  If  a  "boom  man"  were  to  attempt 
to  orient  a  pressure  microphone  so  that  its  zero-degree  incidence  axis 
would  point  to  each  of  the  actors  whenever  he  was  speaking,  several 
"takes"  would  probably  be  required  to  insure  good  intelligibility  in 
the  reproduced  sound,  if  it  can  be  had  at  all.  On  the  other  hand,  if 
the  pressure  microphone  remains  stationary,  with  the  zero-degree 
incidence  axis  pointing  midway  between  the  speakers,  a  type  of 
recording  can  result  that  is  sorely  lacking  at  the  high  frequencies 
because  of  the  narrow  directional  characteristic  that  so  many  pressure- 
operated  transmitters  exhibit..  This  deficiency  in  high  frequencies 
can,  of  course,  later  be  remedied  during  re-recording  by  high-frequency 
equalization  in  the  dubbing  channel.  Such  post-equalization  is,  of 
course,  necessary  only  when  parts  of  the  picture  are  recorded  with 
the  zero-degree  incidence  axis  pointing  directly  at  a  speaker.  If  all 
dialog  is  recorded  at  a  certain  angle  off  this  zero  axis,  only  the  amount 
of  pre-equalization  is  used  in  the  recording  channel  that  will  result 
in  maximum  intelligibility  and  naturalness  of  speech.  Oversight  on 
the  part  of  a  boom  man,  however,  as  well  as  almost  unavoidable 
speaker  configurations,  tending  to  produce  "beam-on"  recording, 
make  such  a  microphone  difficult  to  manipulate,  particularly  when  a 


April,  1939]  UNIDIRECTIONAL  MICROPHONE  TECHNIC 


387 


crew  is  pressed  for  time.  A  unidirectional  microphone,  however,  even 
as  a  velocity  microphone,  preserves  the  balance  between  the  high  and 
low  frequencies  in  a  recording  up  to  very  large  angles  of  reception. 

Large  Solid  Angle  of  Reception. — The  use  of  two  or  more  micro- 
phones to  cover  a  large  scene  has  several  disadvantages.  The  level 
from  the  different  microphones  must  at  all  times  be  balanced  properly 
by  the  mixer,  who  is  already  called  upon  to  control  the  overall  gain  to 
the  amplifiers.  Two  or  more  men  may  be  necessary  to  manipulate 
the  microphones,  causing  additional  difficulties  in  maintaining  a 


FIG.  11. 


Arrangement  of  microphone  and  symphony 
orchestra. 


D,  Director 
M,  Microphone 

B,  4  Bassoons 

C,  4  Clarinets 
F,  4  Flutes 
Hlt  2  Harps 

H2,  8  French  Horns 
Ob,  3  Oboes 
Ti,  3  Trumpets 


TZ,  2  Tympani  and  Traps 

7^3,  4  Trombones 

T4,  1  Tuba 

Vi,  12  First  Violins 

V2,  10  Second  Violins 

V3,  8  Violas 

F4,  6  'Cellos 

V6,  4  String  Bass 

Total:   75  Musicians 


constant  orientation  of  the  microphones  if  they  are  of  the  pressure- 
operated  type.  Interference  effects  caused  by  the  voltages  from  the 
the  microphones  being  out  of  phase  because  of  the  spatial  separation 
of  the  transmitters,  may  produce  a  blurred  or  "bumpy"  recording 
not  always  readily  detected  by  the  mixer.  Fig.  7  shows  how  a  row 
of  soldiers  can  be  covered  well  by  a  single  unidirectional  microphone; 
for  the  sake  of  illustration  there  is  also  indicated  the  directional  re- 
sponse of  two  pressure-operated  microphones  so  placed  to  give  similar 
coverage.  Fig.  8  is  a  similar  application  of  the  unidirectional  micro- 
phone covering  a  large  garden  party  near  a  house  by  a  noisy  street. 


388 


J.  P.  LlVADARY  AND  M.  RETTINGER  [J.  S.  M.  P.  E. 


In  the  ordinary  motion  picture  set  consisting  of  a  floor  and  three 
walls,  the  main  part  of  the  reflected  sound  which  adds  to  the  illusion 
of  an  interior  setting  is  that  reflected  from  the  set  walls.  Reflections 
from  the  absorbent  walls  of  the  sound-stage  are  either  too  weak  or, 
if  noticeable,  are  usually  delayed  too  long  in  time  to  contribute 
pleasingly  toward  such  an  illusion.  Hence,  if  the  atmosphere  of  an 
interior  setting  is  to  be  preserved  within  its  proper  limits,  it  becomes 
necessary  to  collect  as  much  as  possible  these  reflections  from  the  set 
walls  before  they  strike  the  walls  of  the  sound-stage.  In  this  matter 
also  a  unidirectional  microphone  with  its  wide  pick-up  angle  is  an 

effective  device  toward  the  more 
realistic  representation  of  the 
picture  shown  on  the  screen. 

Finally,  when  recording  music, 
a  satisfactory  balance  can  often 
be  achieved  by  the  use  of  one 
unidirectional  microphone  and  a 
judicious  placement  of  the  in- 
strument. Fig.  9  shows  the  re- 
cording of  a  soloist  accompanied 
by  a  piano.  The  distance  be- 
tween the  vocalist  and  the  micro- 
phone should  be  determined  by 
the  strength  of  his  or  her  voice, 
and  the  piano  should  be  placed 
accordingly  for  proper  balance. 
Fig.  10  shows  a  set-up  for  record- 
ing a  small  band,  such  as  a  dance 
orchestra.  The  diagram  is  self-explanatory,  the  only  precaution 
necessary  being  to  keep  the  soloist  at  least  two  feet,  and  preferably 
three,  from  the  microphone.  Fig.  11  shows  microphone  and  orches- 
tra arrangement  for  a  symphony  orchestra. 

There  are  also  situations,  however,  that  lend  themselves  less 
successfully  to  the  use  of  a  unidirectional  microphone.  Such  scenes 
usually  involve  several  actors,  all  of  whom  are  speaking,  while  it  is 
intended  to  record  very  clearly  the  voice  of  only  one.  Obviously,  in- 
stead of  using  a  microphone  having  a  wide  pick-up  angle,  a  transmitter 
with  a  narrow  solid  cone  of  reception  should  be  used.  Likewise,  when 
for  some  reason  or  other  very  close  talking  is  necessary  it  is  advisable 
to  use  a  microphone  that  will  not  cause  an  increase  in  the  low-fre- 


FIG.  12.  Decibels  correction  as  a 
function  of  frequency  and  distance  of 
new  type  unidirectional  microphone 
from  a  point  source  of  sound  (add 
correction  to  plane  wave  calibration  of 
microphone). 


April,  1939]  UNIDIRECTIONAL  MICROPHONE  TECHNIC  389 

quency  response  when  the  distance  between  the  speaker  and  the 
microphone  is  decreased.  This  low-frequency  build-up  is  by  no 
means  so  pronounced  in  a  unidirectional  microphone,  however,  as  it 
is  in  a  velocity  microphone.  Fig.  12  shows  the  variation  of  low-fre- 
quency response  of  a  unidirectional  microphone  with  variation  in 
microphone  distance. 

This  variation  in  low-frequency  response  with  angle  of  incidence 
may  be  calculated  as  follows  :    Let 

Ep  =  Voltage  output  on  a  nondirectional  microphone. 

Cp  =  Sensitivity  constant  of  this  nondirectional  microphone. 

Ev  —  Voltage  output  of  a  bidirectional  microphone. 

Cv   =  Sensivity  constant  of  this  bidirectional  microphone. 

d     =  Microphone  distance. 

F    =  Frequency. 

w     =  2irF. 

A     =  Wavelength. 

t      =  Time. 

6      =  Angle  between  the  direction  of  the  incident  sound  and  the  normal 
to  the  ribbon. 

then      Ep=    C»  sin  wt 
d 


If  Cp  =  Cv  (unidirectional  microphone  having  cardiod  directional 
frequency  response),  then  the  output  of  the  unidirectional  micro- 
phone compared  to  that  of  a  nondirectional  microphone  is  given  by: 


If  0  =  0  (normal  incidence),  this  reduces  to: 

Qi    =   Jl  +  °-006 
\  d2 

or,  expressed  in  decibels: 


REFERENCES 

1  OLSEN,  H.  F.,  AND  MASSA,  F.:  "Applied  Acoustics,"  P.  Blakiston's  Sons  Co., 
Inc.  (Philadelphia),  1934,  p.  141. 

2RETTiNGER,  M.  :  "Note  on  the  Velocity  Microphone,"  J.  Soc.  Mot.  Pict. 
Eng.  XXIX  (Dec.,  1937),  p.  629. 

8  MAXFIELD,  J.  P.:  "Some  Physical  Factors  Affecting  the  Illusion  in  Sound 
Motion  Pictures,"  J.  Soc.  Mot.  Pict.  Eng.,  XVII  (July,  1931).  p.  69. 


ARTIFICIALLY  CONTROLLED  REVERBERATION* 

S.  K.  WOLF** 

Summary. — In  the  recording,  acoustical,  and  electrical  transmission  of  sound,  the 
control  of  reverberation  for  purposes  of  speech  articulation,  musical  quality,  and 
acoustic  illusion  has  been  one  of  the  problems  confronting  architects,  broadcasting, 
recording,  and  acoustic  engineers  for  some  time.  The  paper  discusses  briefly  the 
phenomenon  of  reverberation,  and  describes  two  methods  of  reverberation  control:  (1} 
reverberation  chambers  and  (2)  a  practical  machine  employing  magnetic  recording. 

High-quality  recording  and  reproducing  require  a  wider  and  more 
accurate  control  of  all  acoustic  characteristics,  particularly  reverbera- 
tion, in  the  never-ending  quest  for  perfect  illusion  in  sound  motion 
pictures  and  radio  broadcasting.  Reverberation  chambers  for  add- 
ing a  fixed  amount  of  reverberation  to  a  sound  recording  or  to  a 
broadcast  are  already  being  used  by  the  leading  recording  and  broad- 
casting studios.  These  reverberation  chambers  introduce  an  addi- 
tional liveness  into  the  sound  that  is  not  present  in  the  studio  where 
the  original  sound  is  being  picked  up. 

Such  a  reverberation  chamber  is  normally  a  sizable  room,  approxi- 
mately 10,000  cubic  feet,  with  walls  of  a  glazed  hard-surface  material, 
and  having  a  reverberation  time  of  several  seconds.  In  the  reverbera- 
tion chamber  are  placed  a  loud  speaker  and  a  microphone.  The  loud 
speaker  in  the  chamber  is  connected  by  suitable  means  to  the  micro- 
phone in  the  studio,  and  the  microphone  in  the  reverberation  chamber 
is  electrically  connected  to  the  sound-transmission  channel.  The 
output  of  this  channel  is  then  recorded  or  broadcast.  In  this  way  the 
acoustic  liveness  of  the  reverberation  chamber  will  be  added  to  the 
original  sound  picked  up  by  the  studio  microphone  Such  a  rever- 
beration chamber  offers  only  a  fixed  amount  of  reverberation  and  is 
not  capable  of  instantaneous  variation  or  control. 

The  acoustic  control  of  sound  recording  or  broadcasting  either 
through  reverberation  chambers  or  in  the  space  where  the 

*Presented  at  the  1938  Fall  Meeting  at  Detroit,  Mich.;  received  October  24, 
1938. 
**Acoustic  Consultants,  Inc.,  New  York,  N.  Y. 

390 


ARTIFICIALLY  CONTROLLED  REVERBERATION 


391 


Sound 
Intensity 

A 

(Objective) 


original  sound  is  created,  depends  upon  the  acoustic  properties  of  the 
enclosure  and  the  placement  of  the  microphone  relative  to  the  sound- 
source.  A  correlation  between  microphone  placement  and  acoustic 
characteristics  of  interiors  has  been  worked  out  by  Albersheim  and 
Maxfield.  If  a  sound  is  produced  in  a  room,  a  short  time  elapses 
before  the  intensity  of  the  sound  reaches  a  maximum  value.  This 
time  is  called  the  growth  time.  If  the  sound-source  suddenly  ceases 
to  emit  energy,  time  elapses  be- 
fore the  energy  is  completely 
absorbed  or  falls  below  the 
threshold  of  audibility.  The  re- 
verberation time  is  defined  as 
the  time  required  for  the  sound 
to  diminish  to  one-millionth  of 
its  maximum  value. 

In  Fig.  1  (A)  is  diagram- 
matically  shown  how  the  sound- 
energy  in  a  room  builds  up  to 
the  stationary  value  and  how  it 
decreases  to  the  threshold  of 
audibility.  The  subjective  im- 
pression upon  the  ear,  which  is 

logarithmic,  is  shown  in  Fig.  1  (B). 

Time 


Loudness 

B 

(Subjective) 


FIG.  1.   Diagram  illustrating  growth 
and  decay  of  sound  in  a  room. 


The  growth  time  of  the  sound  is 
scarcely  detected  (curve  B)  be- 
cause subjectively  the  sound 

seems  to  reach  its  peak  value  almost  instantaneously,  whereas  the 
decay  is  more  noticeable  since  it  drops  from  a  higher  to  a  lower  in- 
tensity proportionately  with  time. 

Experience  has  shown  that  music  and  speech  require  different  re- 
verberation times  and  that  different  musical  compositions  require 
different  rates  of  decay.  Since  it  is  always  rather  difficult  to  change 
the  acoustic  characteristic  of  a  room  at  will  there  has  been  for  many 
years  a  need  for  a  practical  method  of  artificially  controlling  rever- 
beration for  producing  any  desired  rate  of  decay  of  the  sound.  Since 
the  growth  of  the  sound  is  scarcely  discernible,  such  a  reverberation 
machine  need  only  control  the  decay  rate.  However,  the  growth  of 
the  sound  may  be  artificially  produced,  if  desired  for  any  reason,  by 
the  same  method. 

The  problem  of  artificial  reverberation  can  also  be  expressed  a  little 


392 


S.  K.  WOLF 


[J.  S.  M.  P.  E. 


differently.  If,  by  some  artificial  means,  the  intensity  of  a  sound  can 
be  controlled  or  regulated  so  as  to  correspond  to  the  natural  change  of 
intensity,  the  effect  must  be  the  same  as  the  effect  of  natural  rever- 
beration. 

A  simple  way  of  producing  reverberation  artificially  is  to  record  the 
sound  and  to  have  a  number  of  time-displaced  pick-up  heads  feed  the 
energy,  through  adjustable  volume  controls,  back  into  the  trans- 
mission line  to  which  the  microphone  is  connected,  through  an 
amplifier  of  the  desired  gain  and  frequency  characteristics. 

The  volume  controls  are  so  adjusted  that  the  amounts  of  energy 
supplied  to  the  transmission  line,  from  one  pick-up  to  the  next,  follows 
an  exponential  law.  To  approach  theoretically  the  natural  rate  of 


FIG.  2.     Circuit  diagram  of  reverberation  unit. 

decay,  an  infinite  number  of  pick-up  heads  would  be  required.  For 
most  practical  purposes,  however,  several  reproducing  heads  are 
sufficient. 

An  artificial  reverberation  machine  must  meet  the  following  re- 
quirements : 

(1)  Immediate  playback. 

(2)  Low  maintenance  cost. 

(3)  Control  of  intensity  and  frequency  response. 

(4)  Mechanically  and  electrically  fool-proof. 

Existing  methods  of  sound  recording  can  not  fulfill  these  require- 
ments. Film  recording  used  in  the  sound  motion  picture  industry 
requires  processing  before  reproducing.  Mechanical  recording  is  too 
expensive  because  of  the  necessity  of  continuously  using  new  record- 
ing material.  There  are  other  methods  of  recording  that  are  a  prac- 


April,  1939]         ARTIFICIALLY  CONTROLLED  REVERBERATION 


393 


tical  solution  of  the  problem.  One  of  the  methods  makes  use  of  a 
tape  coated  with  fluorescent  material,  which,  when  exposed  to  ultra- 
violet light  controlled  by  a  suitable  light- valve,  will  fluoresce  for  a 
short  while.  This  tape  passes 
a  number  of  photocells,  which 
then  pick  up  the  sound  and  re- 
produce it.  The  exponential 
decay  of  the  fluorescence  has  at- 
tracted experimenters  to  this 
means  of  creating  artificial  rever- 
beration. After  a  certain  time 
the  emission  ceases,  particularly 
if  the  exposure  was  made  to 
infrared  light,  and  a  new  re- 
cording can  take  its  place.  In 
this  method  the  background- 
noise  remains  constant  as  the 
signal-level  diminishes;  the  sig- 
nal-to-noise ratio  therefore  de- 
creases. For  long  periods  of 
reverberation  the  fluorescence 
decays  too  rapidly. 

The  author's  method  makes 
use  of  the  principle  of  magnetic 
recording  that  has  proved  prac- 
ticable. A  magnetic  record  may 
be  instantly  reproduced  and  im- 
mediately obliterated.  A  com- 
plete description  of  the  principle 
of  magnetic  recording  has  been 
published  in  the  JOURNAL  by 
S.  J.  Begun.1  For  artificial  re- 
verberation the  tape  may  be 
used  as  shown  in  Fig.  2.  The 
recording  head  a  is  supplied 
with  energy  from  the  microphone  b  and  amplifier  c.  The  pick-up 
heads  (d,  e,  f,  and  g)  and  the  recording  amplifier  are  connected  to  a 
mixer  amplifier,  the  output  of  which  can  either  feed  into  a  trans- 
mission line  or  a  loud  speaker.  The  mixer  amplifier  is  provided 
with  adjustable  filters  for  the  pick-up  heads  to  make  it  possible  to 


FIG.  3. 


Mechanism  for  driving 
the  tape. 


394 


S.  K.  WOLF 


[J.  S.  M.  P.  E. 


get  any  frequency-response  desired  in  reproduction.  Such  adjust- 
ment of  response  is  valuable  since  the  reverberation  time  is  not 
the  same  for  all  frequencies,  and  depends  upon  the  absorption  coeffi- 
cient of  the  room  for  different  frequencies.  It  is  very  important 
that  the  sound-carrier  move  without  appreciable  variation  of  speed, 
particularly  since  the  machine  is  used  most  frequently  for  musical 
reproductions.  Fig.  3  shows  a  mechanism  for  driving  the  tape.  A 
motor  drives,  through  a  belt,  one  cylinder  a,  and  additional 

cylinders  b  are  driven  by 
the  endless  tape  loop  c,  the 
ends  of  which  are  joined  over 
two  cross-over  rollers  d.  This 
endless  loop  can  be  made 
long  enough  for  any  desired 
reverberation  time,  and  as 
many  reproducing  heads  can 
be  arranged  as  required.  The 
machine  is  normally  provided 
with  eight  reproducing  heads 
logarithmically  displaced  in 
time;  the  first  so  located  that 
it  will  pick  up  the  record  af- 
ter 0.1  second,  the  second 
after  0.16  second,  the  third  af- 
ter 0.25  second,  the  fourth 
after  0.4,  the  fifth  after  0.66, 
the  sixth  after  1  second,  the 
seventh  after  1.7  seconds, 
the  eighth  after  2.7  seconds. 
The  arrangement  of  the  com- 
plete unit  for  motion  picture  studio  and  broadcasting  use  is  shown  in 
Fig.  4.  The  equipment  is  rack  mounted,  enclosed  in  a  cabinet,  and 
supported  on  casters  so  that  it  can  be  easily  moved  from  one  studio 
to  another. 

Magnetic  recording  meets  the  quality  requirements  for  synthetic 
reverberation :  Its  frequency  responses,  using  adequate  equalizers,  are 
flat  from  50  to  7500  cycles,  and,  if  desirable,  even  higher.  The 
signal-level  is  40  db.  above  the  background-noise.  A  sound-carrier  of 
steel  tape  0.1200  mil  wide  and  0.003  mil  thick  is  used.  This  steel  tape 
is  particularly  strong,  and  tests  have  shown  its  mechanical  strength 


10.5" 

\ 

9" 

o 

0 
O 

0 

o 

0 

RF.PRODUCIN&  Annif/fR 

0 
0 

0 

I  [S]  ©@© 

Pil/6S                           Si' 

Swee  PAUCL             as" 

Jj" 

VowHcCo/anoi  OF  PICK  UP  Hc*as\- 

o 

0 

°©@©@© 

JhflBE. 

3.5 

o@©©@© 

0 
0 

o 

K 


FIG.  4. — Arrangement  of  reverberation 
control  unit. 


April,  1939]         ARTIFICIALLY  CONTROLLED  REVERBERATION  395 

and  electromagnetic  properties  to  be  entirely  satisfactory.  The  ma- 
chine requires  little  servicing  and  is  simple  to  operate.  The  mecha- 
nism shown  in  Fig.  3  has  been  designed  jointly  by  S.  J.  Begun, 
A.  Stapler,  and  the  author.  The  electrical  and  physical  designs  of  the 
unit  have  been  anticipated  by  A.  N.  Goldsmith2  in  his  patent  on  syn- 
thetic reverberation  under  which  this  company  is  licensed. 

REFERENCES 

1  BEGGUN,  S.  J. :     "Recent   Developments   in   Magnetic   Sound   Recording," 
J.  Soc.  Mot.  Pict.  Eng.,  XXVIII  (May,  1937),  p.  464. 

2  GOLDSMITH,  A.  N. :    U.  S.  Pat.  2,105,318. 

DISCUSSION 

DR.  GOLDSMITH  :  The  method  here  described  is  obviously  one  of  which  we  shall 
hear  more  as  time  goes  on,  because  it  is  impossible  that  reverberation  of  all  sorts 
and  types  will  be  produced  merely  by  simulating  the  actual  physical  chamber  or 
enclosure  in  which  the  original  reverberation  might  take  place.  It  is  manifestly 
not  economical  in  the  long  run  to  have  a  great  number  of  empty  rooms  or  pipe 
lines  of  different  characteristics  not  flexibly  controllable,  when  such  an  ingenious 
mechanism  as  Mr.  Wolf  has  described  can  be  conveniently  and  economically  used 
for  the  purpose. 

The  versatility  of  that  type  of  mechanism  depends  upon  the  work  in  connection 
with  magnetic  recording  that  was  described.  The  problem  of  producing  a  reliable 
telegraphone  is  not  so  simple  as  it  sounds,  because  if  one  takes  a  piece  of  steel 
wire  and  tries  to  run  it  around  in  a  loop  over  and  over  again,  he  soon  makes  the 
interesting  discovery  that  steel  can  be  defined  as  a  material  that  kinks  and  snaps 
itself  automatically.  The  problem  of  producing  telegraphones  with  wire  elements 
is  difficult  if  one  requires  reliable,  continued  operation,  and  it  takes  a  clever  ar- 
rangement, such  as  this  steel-tape  holder,  to  accomplish  the  results  in  heavy-duty 
service. 

The  flexibility  of  the  circuits  and  arrangements  that  have  been  shown  is  perhaps 
greater  than  has  been  indicated.  Thus,  if  one  wants  to  simulate  the  acoustics  of 
a  room  which  has  walls,  floor,  and  ceiling  of  different  characteristics,  one  can 
divide  the  reproducing  pick-up  heads  into  three  major  groups  corresponding 
respectively  to  the  horizontal  right  and  left  walls,  the  front  and  back  walls,  and 
the  ceiling  and  floor,  and  then  put  into  the  circuits  of  the  reproducing  heads  cor- 
rective networks  corresponding  in  general  to  the  frequency  and  also  the  phase-of- 
reflection  characteristics  of  the  walls  in  question;  so  that  one  can  modify  each  of 
the  three  groups  of  major  reflections  more  or  less  systematically  as  desired. 

Other  interesting  possibilities  exist.  In  the  panel  arrangement  that  was  shown 
are  a  group  of  reproducer  amplitude  controls — that  is,  volume  controls.  It  is 
possible  to  interconnect  those  by  chains,  levers,  or  other  mechanical  means,  so  as 
to  have  a  master  control.  One  can  even  move  the  recording  heads  along,  changing 
the  time  of  each  delay ;  or  vary  their  amplitude  systematically,  or  both,  so  that  by 
means  of  a  simple  knob  one  can  go  to  extremes  of  flexible  control,  one  can  change 


396  S.  K.  WOLF  [j.  s.  M.  p.  E. 

the  delays  in  each  reproducer  head  relative  to  fundamental  sounds  and  change  the 
relative  amplitudes  and/or  tone  qualities  from  each.  So  one  could  conceivably 
have  a  knob  with  a  pointer  moving  over  a  scale  marked:  "small  room  of  wood," 
"larger  room  of  stone,"  "scene  under  a  bridge,"  "cathedral  interior,"  and  so  on; 
and  by  merely  turning  the  knob  one  could  actually  change  the  resulting  acoustics 
over  this  wide  series  of  conditions. 

There  are  other  interesting  applications,  as  Mr.  Wolf  has  pointed  out,  for  this 
technic ;  but  one  of  the  main  points  that  should  be  considered  is  that  the  acoustic 
output-to-ground-noise  ratio  of  each  of  the  echoes  produced  by  the  machine  re- 
mains satisfactory. 

There  is  another  point  of  interest.  It  is  believed  telegraphone  recording  equip- 
ment is  entirely  adequate  for  the  purpose  mentioned,  for  two  reasons,  first,  be- 
cause its  tonal  quality  is  very  high  when  it  is  properly  built ;  and,  second,  because 
the  role  played  by  the  very  high  frequencies  is  less  in  synthetic  reverberation  than 
in  the  original  sound. 

Finally,  there  is  another  point  of  interest.  To  the  practical  man  carrying  out 
recording  in  studios,  this  device  offers  the  possibility  of  later  sound  editing  on  a 
large  scale,  because  he  can  take  an  original  record  of  sound  in  "dead"  surround- 
ings, and  then  he  can  add  any  desired  type  of  reverberation  at  any  time  thereafter. 
This  he  can  re-record  in  any  desired  fashion  as  often  as  desired,  and  experiment 
and  produce  different  types  of  reverberation  until  he  finally  hits  the  one  that 
gives  him,  for  example,  the  effect  of  a  man  walking  out  of  a  room  into  a  tunnel  and 
into  a  larger  room.  That  can  be  experimented  with  over  and  over  again  until 
the  desired  result  is  obtained,  without  injury  to  the  original  record. 

Perhaps  I  should  apologize  for  the  length  of  this  discussion,  but  I  am  greatly 
interested  in  the  whole  philosophy  of  producing  sound  effects  by  electrical  means 
rather  than  by  clumsy,  large,  and  costly,  mechanical  means. 

MR.  KELLOGG  :  My  first  SMPE  paper  was  read  in  1928  under  the  title,  "Some 
New  Aspects  of  Reverberation,"  in  which  I  indulged  in  the  speculation  that  since 
we  did  not  need  reverberation  to  help  intensity  any  more  with  our  electrical  appa- 
ratus, we  could  control  things  beautifully  if  we  could  only  get  rid  of  the  natural 
reverberation  and  supply  just  what  we  wanted  where  we  wanted  it  and  when  we 
wanted  it  by  various  electrical  devices.  It  seemed  that  the  remarkable  flexi- 
bility and  power  our  new  electronic  and  acoustic  tools  were  going  to  give  us  better 
control  than  we  have  ever  had  before. 

Well,  I  have  been  waiting  during  the  intervening  ten  years  to  see  any  of  that 
really  done.  After  all,  I  do  not  think  that  such  mere  speculation  as  I  indulged  in 
is  a  particular  credit  compared  to  really  doing  something  about  it. 

There  is  one  aspect  in  the  production  of  reverberation  by  a  recording  sound 
system  which  is  not  necessarily  an  indication  of  its  being  faulty,  but  it  is  funda- 
mentally different  from  what  we  will  get  with  any  normal  natural  reverberation. 
If  the  absorbent  qualities  of  the  walls  are  such  as  to  cause,  let  us  say,  twice  as  much 
absorption  at  5000  cycles  as  we  have  at  2000,  the  2000-cycle  tone  would  be  audible 
for  twice  as  long  a  time  as  the  5000-cycle  tone.  In  the  case  of  reverberation  pro- 
duced by  recording  the  sound,  if  the  rate  of  attenuation  is  determined  by  the  gain 
settings  of  the  several  pick-ups,  all  the  components  will  have  the  same  reverbera- 
tion time.  Therefore,  the  only  kind  of  room  this  will  truly  simulate  is  one  which 
has  exactly  equal  attenuation  at  all  frequencies. 


April,  1939]         ARTIFICIALLY  CONTROLLED  REVERBERATION  397 

I  am  not  saying  that  the  kind  of  reverberation  you  may  get  by  the  recording 
method  may  not  be  even  better,  provided  you  keep  the  ground  noises  down,  but  it 
is  essentially  different. 

There  is  one  phase  of  the  question  that  I  feel  deserves  much  consideration. 
It  seems  to  me  to  be  very  desirable,  if  we  have  a  ready  means  of  producing  arti- 
ficial reverberation,  to  do  it  in  the  listening  and  not  in  the  recording.  I  grant  that 
by  doing  it  in  the  recording  you  make  it  available  for  all  theaters,  but  the  ideal 
system  which  I  hope  ultimately  to  see  in  use  is  one  in  which  the  original  sound 
recorded  with  very  little  reverberation,  and  in  the  reproduction  the  non-reverber- 
ant sound  will  come  from  a  speaker  directly  in  front  of  you,  and  echoes  will  come 
from  speakers  scattered  around  the  room. 

We  tried  some  experiments  a  number  o